The bows of the South Andaman group of tribes are always decorated with

incised patterns. The conventional pattern is shown in Fig. 10. One

line of such pattern runs down each edge of both the inside and the

outside of the bow, and on the inside a similar line of pattern runs

down the middle. When bows are newly made they are often also decorated

with designs in red paint and white clay, particularly if they are

intended as gifts. These painted designs soon wear off and are not

renewed.

In these tribes bows are sometimes made of a size so large as to be

almost useless for hunting. One such bow, now in the Cambridge Museum,

is 220 cm. long and with a maximum breadth of 10 cm. Such bows are very

carefully made and decorated and are intended as gifts. A man generally

makes such a bow with the deliberate intention of giving it to some

person whom he wishes to please. The bow that I have was specially made

to give to me in this way. A man who possessed such a bow would not

dream of using it in hunting, but he might use it in a shooting match,

in order to show his skill.

In the South Andaman tribes toy bows are made for boys of somewhat the

same shape as the ordinary hunting bow. An example of such a bow, now

in the Cambridge Museum, is 121 cm. long. It is cut from a bent piece

of wood in such a way that the lower portion is curved outwards. The

section in the middle is plano-convex, very nearly the half of a

circle, the breadth being 26 mm. and the thickness 13 mm. It is

broadest in the middle, and tapers towards each end. When strung it

assumes the typical S-shape of the South Andaman bow. It is served over

with thread at one end and with a strip of cotton cloth at the other,

leaving two points for the string. The latter is of the usual South

Andaman type, but of smaller dimensions.

We must turn now to the bow used by the four tribes of the North

Andaman, which is of a somewhat different pattern from that just

described. It is, in the first place, shorter, lighter and more

slenderly made. Of ten typical specimens the shortest is 153 cm. long,

and the longest is 182 cm. The usual length is about 160 to 165 cm. In

its broadest part the North Andaman bow is broader than the South

Andaman bow, the breadth varying from 6·5 to 7·5 cm. in different

specimens.

Although the North Andaman bow is, as a rule, cut from a curved piece

of wood, it may, on occasion, be cut from a piece that is practically

straight.

At the upper end of the bow there is a long point. In a specimen that

is in every way typical, at about 5 cm. from the point the section is

circular and the diameter is about 5 mm.; at about 30 cm. from the end

the section is slightly flattened or oval, and the maximum diameter

(the breadth from side to side) is about 1·5 cm. From this the bow

broadens fairly rapidly, until at a distance of about 60 cm. from the

end it is 7 cm. broad. The section at a point 60 cm. from the end is

shown in Fig. 11, where it may be seen to be convex on the outside and

only very slightly convex on the inside. At about 75 cm. from the end

the bow narrows in breadth to form a handle, at the same time

increasing slightly in thickness. The handle is approximately circular

in section in the middle, which is about 95 cm. from the upper end, and

about 80 cm. from the lower end, the diameter being about 2·2 cm. Below

the handle the bow broadens out once more into a lower blade which in

shape and section is similar to the upper blade. At a distance of about

30 cm. from the lower end the bow once more narrows off to a point

approximately circular in section. The lower point of the bow is not so

long or so tapering as the upper point.

The whole bow thus consists of two blade-shaped portions tapering to a

point at each end, and with a waist or handle between them. The upper

blade is straight, i.e. is not curved either outwards or inwards. The

lower blade is curved outwards (like that of the South Andaman bow) in

nearly every newly made bow and in every bow that has been in use.

The upper part of the bow is served over with string for about 1·5 cm.

(at a distance of 15·5 cm. from the end in the bow that has been

described), and the lower end is similarly served (at a distance of 6

cm. from the end). The general shape of the bow as seen from inside is

shown in Fig. 12.

The bow-string of the North Andaman is made from Anadendron fibre in

much the same way as described in connection with the South Andaman

bow, but in the North there is a loop or eye at both ends of the

string. As soon as the first few centimetres of the cord have been made

(by the method previously described) it is bent over into a loop, and

this loop is served with Anadendron thread, just as in the case of the

South Andaman string. The making of the string then proceeds in the

usual way until a sufficient length has been made, this depending, of

course, on the length of the bow for which it is intended. The end is

then bent over into a loop, and this loop is served over with thread.

The loose end of fibre is not in this case (as it is in the South

Andaman string) twisted round the standing part of the cord, but is

laid beside it, and the thread that has been used for serving is wound

spirally round them both for a distance of about 10 cm. from the neck

of the loop, so that the end is stopped.

When the bow is to be strung the first made loop is slipped over the

top end of the bow so that it rests on the thread serving already

mentioned, the neck of the loop being on the inside of the bow. The bow

is then laid on the ground, inside downwards, a foot is placed on the

middle, and the lower end of the bow is bent upwards (and therefore

outwards) far enough to allow the other loop of the cord to be slipped

over the end. The bow is now in what may be called the half-strung

position, and in order to understand the mechanical principles of this

type of bow it is necessary to make quite clear what this position is.

It is shown in Fig. 13, A. The string passes from the top to the bottom

on the outside of the bow, so that the bow is, so to speak, reversed,

and is subjected to a strain that causes it to curve outwards. In most

bows, when they are first made, there is an outward curve in the lower

portion, owing to the bow having been cut from a curved piece of wood.

When the bow is half strung this outward curvature is increased. If a

bow be made from a straight piece of wood, an outward curve is produced

by the operation of stringing it, as described above.

As soon as a bow is completed it is strung in the reversed position

described, and is then placed over a fire, in such a position that the

lower (curved) blade is immediately above the fire. The smoke and heat

of the fire season the wood of this portion of the bow. Any specimen of

a bow of this type, unless it has been newly made and not seasoned, is

blackened on the inner surface of its lower part. The bow is left to

season in this way for some time. A man places his bow over the fire of

his own hut, which is kept constantly burning day and night. It must be

remembered that all the time it is being seasoned the bow is subjected

to a slight strain curving it outwards.

After the bow has been sufficiently seasoned it is brought into use.

When a bow that is half strung or strung in a reversed way is to be

used, it is taken by the handle in the left hand, with the string away

from the body, the bow being upside down. The lower part of the bow

(i.e. what is really the top of the bow when it is in its normal

position) is rested against the thigh. The string is taken in the right

hand and pulled over towards the body, so that the bow reverses itself

and appears in the fully strung position shown in Fig. 13, B. It is

then ready for immediate use.

A bow of this type is hardly ever entirely unstrung. When a man has

finished with his bow for the time being, he puts it once more in the

half-strung position, by an action the reverse of that described above,

and then hangs the bow over the fire. Thus while the bow is in active

use it is in the fully strung position, and at all other times it is

kept in the half-strung position.

It is clear that the North Andaman bow depends on a principle that is

not made use of in the South Andaman bow, which we may state by saying

that if a piece of wood be subjected to the influence of heat and smoke

while it is bent in one position it will acquire greater strength and

elasticity to react against a strain that bends it in the opposite

direction. When the bow is fully strung it is S-shaped. When it is

drawn the greater part of the strain falls on the lower portion where

it is curved outwards. It is this portion of the bow that is

strengthened by seasoning.

The North Andaman bow is very much lighter than the South Andaman bow

and is much more elastic. I always found it very difficult to shoot

with a South Andaman bow, but on the other hand I found the North

Andaman bow very easy to use. In drawing it only a slight pull is

required in order to send an arrow with considerable velocity. The

disadvantage of the northern bow is that, owing to its slighter build,

it does not last very long, and is liable to be broken. However, it

only takes a man a few days to make a new bow, string included, and the

very definite superiority of the North Andaman bow over that of the

South Andaman amply compensates for its shorter life.

Bows of exactly the same shape but of smaller dimensions are made in

the North Andaman for boys, the length varying from about 90 cm. to

about 120 cm. For very small boys toy bows of a different pattern are

made. The bow is formed of a piece of wood about 90 cm. long and from 2

to 2·5 cm. broad in the middle. The section in the middle is

convexo-convex, with a high degree of convexity on the outside and a

much slighter convexity on the inside. The bow tapers to a point at

each end, but it tapers more gradually at the top than at the bottom.

The bow-string is a simple piece of string (two-ply) made of Anadendron

fibre. It is tied to the lower end of the bow at a distance of 1·5 to 3

cm. from the end, and at the top it is tied at from 4 to 7 cm. from the

end. The shape of a toy bow of this kind as seen from the side is shown

in Fig. 14. It is not S-shaped, like the toy bow of the South Andaman

previously described, but the curvature is asymmetrical.

I obtained a specimen of a toy bow made of bamboo. Unfortunately there

was no string, but it was probably intended to be strung in the fashion

of the North Andaman toy bow just described. The outer surface of the

bamboo was the outside of the bow, with the result that in section the

inner side of the bow was more convex than the outer side. This is the

only bow in the whole of the Andamans in which I ever saw this feature.

In all other bows of whatever type the outer side is markedly convex

and the inner side is either concave, flat or only slightly convex.

It is now possible to compare one with another the different forms of

bow in use in the Andamans. It would seem almost certain that the North

Andaman bow can only have been derived from the form in use in the

South Andaman or from one very similar to it. It is only after they

were in the habit of making bows with an outward curve in the lower

portion that the natives could have devised the method of seasoning

this portion of the bow and keeping it in the reversed or half-strung

position. It is unnecessary to argue the matter in detail, and we may

conclude that the North Andaman type is derived from the South Andaman

type.

It is less certain, but still highly probable, that the South Andaman

form was derived from a bow similar to that still in use in the Little

Andaman. The South Andaman toy bow shows a stage intermediate between

the Little Andaman bow and the usual South Andaman form. The section of

this toy bow is very similar to that of the Little Andaman bow. It has

no blades, and therefore no waist for the handle. The shape, however,

is asymmetrical. Owing to the different method of stringing it, the

shoulder at the end of the bow is unnecessary, and the bow is

strengthened (prevented from splitting so easily) by tapering the end

to a point instead. The difference between the toy bow and the general

South Andaman bow is the presence in the latter of the two blades and

the waist. The broadening of the bow into the blades is necessary in

order to prevent it from accidentally reversing itself.

We have still to consider the modified form of the Jarawa bow. The

origin of this is easy to discover by the examination of a few typical

specimens. Since the Jarawa have been in the South Andaman they have

been in hostile contact with the tribes of the South Andaman Division.

They have had opportunities of handling bows of the kind made by these

tribes, and they have apparently discovered that these bows are more

efficient than their own, but they have had no opportunity, such as

only friendly intercourse would give, of discovering the principles on

which the South Andaman bow is made. They have attempted to imitate it

to the best of their ability, and this they have done (1) by making

their bows longer and broader, (2) by making them concavo-convex in

section instead of convexo-convex, (3) by cutting them occasionally

from wood that gives them an outward curve in the lower portion, (4) by

imitating the shape of the handle without, however, giving the bow a

waist, (5) by serving over the bow-string with thread at the point

where the arrow touches it, (6) by ornamenting their bows with incised

and painted patterns. These are the only differences between the

modified Jarawa type and the Little Andaman type, and all these may be

explained as attempts to imitate the bows of the South Andaman

Division. In not a single one of the modified Jarawa bows that I have

seen is the fundamental principle of the South Andaman bow successfully

applied.

We may conclude that the ancestors of the Andamanese at one time used a

bow resembling that in use in the Little Andaman at the present day,

and that the S-shaped bow of the Great Andaman has been invented since

the separation of the two main divisions of the race.

It is not possible, owing to lack of sufficient information, to

determine exactly the types of bow used by the Semang. The following

notes are based on only six specimens, two of which are in the Museum

of Ethnology at Cambridge [249], while the others are in the British

Museum. Four of the specimens are sufficiently similar to one another

to be regarded as belonging to one type, which seems to be the usual

type of Semang bow. In length they are 165, 174·5, 182 and 197·5 cm.,

and at the middle they vary in breadth between 2·5 and 3 cm. Three are

of palm-wood, and the other is of a light-coloured but tough kind of

wood. The shape of the section in the middle of the bow varies

considerably in the different specimens, as shown in Fig. 15. There is

some uncertainty as to which is the inside, and which is the outside of

the bow. At each end there is a shoulder, the point or tip at one end

being in every instance considerably longer than the other. The string

in each case is a three-ply rope, having a spliced eye at one end and

being fastened to the bow with a knot at the other end.

The other two specimens do not conform to this type. One is of wood,

134 cm. long, and 3·5 cm. broad in the middle. One end is provided with

a notch on each side for the string, while the other end has three

pairs of notches. The string is a three-ply cord with a loop or eye at

each end. The other specimen is made of bamboo, and is 147 cm. long and

2·25 cm. broad in the middle. One end has one pair of notches and the

other has two pairs. The string is two-ply with a loop at each end.

What we may perhaps regard as the usual type of Semang bow thus differs

from the bow of the Little Andaman in three important respects, (1) in

having a longer point at one end than at the other [250], (2) in having

a string of three-ply rope instead of a strand of twisted fibre, and

(3) in the variations in the section at the middle. From the four

specimens available it is not possible to determine around what norm or

norms the section of different specimens varies.

Turning to the Negritos of the Philippine Islands, although a number of

bows have been described by Meyer [251], the available information is

not sufficient to enable us to determine what are the different types,

and what is their relation one to another. Amongst the different

varieties of bow used by the Negritos there is one which is very

similar to the bow of the Little Andaman. It has a rounded or convex

outer side, and a flattened inner side. The string loops on to a point

at each end, and the string itself is formed of twisted fibre.

From a study of the available material it seems that we are justified

in concluding that the primitive Negrito bow was made of wood, with a

shoulder at each end, probably with a section rounded on one side and

flattened or keeled on the other, and that it had a string of twisted

fibre with a loop or eye at one end, the other end being attached with

a knot.

The Andaman Islanders use two different kinds of arrows, one for

shooting fish and the other for shooting pigs.

The common fish-arrow, as at present used in all parts of the Great

Andaman Division, consists of three parts—a shaft, a fore-shaft, and a

point. The shaft is a length of bamboo straightened by means of heat,

and may vary in length from 70 to 110 cm. At one end the bamboo is cut

off about 3 cm. beyond one of the nodes, and a nock is made (Fig. 16).

At this end the shaft is roughened with a Cyrena shell so as to give a

firm grip for the fingers. At the other end the shaft is tapered for

about a centimetre. At this end the bamboo is roughened with a shell to

give a hold to the thread with which it is bound. The fore-shaft is a

length of wood from 15 to 40 cm. long. One end is slightly tapered so

that it fits tightly into the end of the bamboo shaft. The other end is

tapered to a point, which is flattened on one side. The fore-shaft is

inserted into the end of the shaft and the joint is bound over for a

distance of about 2·5 cm. with thread. The point consists of a piece of

iron wire, sharply pointed at both ends. This is laid against the

flattened side of the fore-shaft and is bound to it with thread in such

a way that one point projects at the back to form a barb (Fig. 16).

When the binding is completed it is covered entirely over with a

composition made by melting together bees’-wax, resin and red ochre.

The composition is melted over a fire, is applied with a short piece of

wood and is then smoothed over with a hot Cyrena shell. Only the

binding attaching the point to the fore-shaft is covered with

composition, and not that at the joining of the fore-shaft and the

shaft.

Arrows of this kind are used for shooting fish, but they also serve to

shoot snakes or rats and on rare occasions birds.

Similar fish-arrows are made in the Little Andaman, but they are larger

(i.e. both longer and thicker) than those of the Great Andaman

Division. The two ends of the bamboo shaft are not roughened, and the

binding attaching the point to the fore-shaft is not covered with

composition (which seems to be unknown in the Little Andaman) but with

bees’-wax only.

This seems to be the traditional form of fish-arrow of the Andamans.

Before iron was plentiful the point consisted either of the serrated

bone from the tail of the sting-ray or of a piece of the tibia of a pig

ground down to the requisite dimensions on a piece of stone, and

sharpened at each end.

In the Little Andaman fish-arrows are sometimes used with two or four

prongs attached to a bamboo shaft. In the British Museum there is an

arrow from the North Sentinel Island with four prongs tied on to a

wooden shaft, each prong being barbed by a detached piece of wood at

the end.

A simple form of arrow is made in both the Great Andaman and the Little

Andaman consisting of a bamboo shaft with a pointed wooden head, the

point being hardened in the fire. Such arrows are now very rarely used,

save for shooting at a mark, but it is probable that before iron was

plentiful they were used as a substitute for the fish-arrow described

above, being easier to make although less serviceable.

The pig-arrow in use in the Great Andaman consists of a shaft, and a

fore-shaft to which is attached a head. The shaft is a piece of the

wood of the Tetranthera lancæfolia, cut from the tree and straightened

by means of heat. At the narrower end a nock about 1 cm. deep is cut,

and the arrow is served with thread of Anadendron fibre for about 1·6

cm. above the nock, in order to prevent splitting. At the other end the

shaft is hollowed out to a depth of about a centimetre. This hollowing

is done with the point of a fish-arrow or other similar piece of

pointed iron. For a distance of about 1·5 cm. the end of the shaft is

served over with thread of Anadendron fibre, so as to prevent it from

splitting.

The fore-shaft consists of a piece of tough wood one end of which is

cut to such a size that it will fit fairly tightly into the hollow at

the end of the shaft. At the other end it is split so as to admit the

head.

The head consists of a piece of iron broken into shape with the aid of

a stone hammer and then ground down and sharpened on a whet-stone or

with a file if one be obtainable. The usual shape is shown in Fig. 17.

The head is inserted into the split end of the fore-shaft and the end

of the latter is then served over with thread. A few centimetres below

the head either one or two sharp-pointed pieces of iron wire are placed

against the fore-shaft in the same plane as the head and are bound

firmly to it with the thread, so as to provide a barb or barbs.

A cord is made and one end of it is attached to the shaft and the other

end to the fore-shaft. This cord is made as follows. A number of

strands of Anadendron fibre are taken and waxed with bees’-wax. These

are made into a cord by the same method as that described in connection

with the Great Andaman bow-string, one strand being wrapped round the

others. About 40 cm. of single cord is made in this way and the two

ends are tied together. A piece of elastic wood is bent into the form

of an arc and the loop of cord is placed over this so that it is

stretched tight. A length of thread (of Anadendron fibre) is made and

wound on to a fine netting needle or on to a thin slip of wood, and the

two cords as they are stretched side by side are bound together with

this thread by the process known as “nippering” (Fig. 19). In this way

a firm and strong flattened cord is produced. One end of this is

fastened to the fore-shaft immediately above the end that fits into the

shaft. The other end is fastened to the shaft a few centimetres from

the end, leaving a short length of the shaft around which the cord may

be spirally wound when the arrow is in use. (Fig. 18.)

Finally, the thread on the fore-shaft, i.e. that which holds the head

and barbs in place, and also that which serves to attach the end of the

cord, are covered with the composition already described as used on

fish-arrows.

Arrows of this kind are used in pig-hunting. The fore-shaft is inserted

into the end of the shaft, the cord attaching the two being wound

spirally round the end of the shaft. When a pig is struck the barbs

prevent the head from coming out of the wound. As the pig attempts to

run away the shaft catches against the undergrowth of the jungle and

comes loose from the fore-shaft. Sooner or later the shaft becomes

entangled in the undergrowth and holds the wounded pig fast till the

hunters can come up with it and despatch it. It is obvious that the

cord of the arrow needs to be so strong that the pig cannot break it.

The natives of the Great Andaman tribes say that before they had plenty

of iron they made similar pig-arrows with heads of shell and barbs of

pig’s bone.

The natives of the Little Andaman make a pig-arrow very similar to that

of the Great Andaman tribes, but on the average somewhat longer. The

cord attaching the fore-shaft to the shaft consists of a length of

double two-ply rope of Hibiscus fibre. The binding of the arrow is done

with thread of the Gnetum fibre instead of Anadendron, and is not

coated with composition but is smeared with bees’-wax. The barb (there

being usually only one) is not fixed in the same plane as the head, but

in the plane at right angles to it. (Fig. 20.)

Amongst the Jarawa the head of the pig-arrow is attached to the

fore-shaft by a different method, holes being made in the iron through

which the thread that holds it is passed. (Fig. 21.)

Another kind of pig-arrow is sometimes made in the Great Andaman that

has not a detachable head. The shaft is a length of bamboo into one end

of which is fitted a fore-shaft of wood. The end of this fore-shaft is

split and a head of iron is inserted into it and bound there. Such an

arrow may be without barbs or may have one or two barbs of iron. It is

used only rarely and then chiefly for despatching a pig that has

already been struck by an arrow of the usual kind. The natives say that

in former times arrows of this kind were used in fighting in preference

to ordinary pig-arrows, which, however, were also used.

In former times the natives of the Great Andaman, according to their

own statements, made an arrow consisting of a bamboo shaft at the end

of which was inserted a head made of Areca wood. An arrow of this kind,

made for me by a native, is shown in Fig. 22.

None of the arrows made in any part of the Andamans is feathered.

In a comparison of the arrows of the Andamanese with those of the

Semang and the Philippine Negritos the most interesting point is that

all three branches of the Negrito race use arrows with detachable

heads. Arrows of this kind from the Philippines are described by Meyer

[252].

An important point of difference would seem to be that while the

Andamanese do not feather their arrows the Semang and the Negritos of

the Philippines do so. It would seem, however, from the account of

Semang arrows given by Skeat [253] that the feathering is such that it

is of no actual service in directing the flight of the weapon, and that

it is doubtful whether the Semang really understand the principle of

feathering, or whether they do not employ it as the mutilated survival

of more intelligent methods or perhaps make use of it for solely

magical purposes.

Another important point of difference is that the Semang poison their

arrows, while the Andamanese do not. In this connection it must be

remembered that the former people have for long been neighbours of

people who use blow-pipes with poisoned darts.

After the bow and arrow the most important hunting weapon of the Great

Andaman is the harpoon which is used in capturing dugong, turtle,

porpoise, and large fish. The harpoon consists of a head, a shaft and a

line. The line is a length of rope of Hibiscus fibre of as much as

twenty fathoms or more in length. The shaft is a bamboo of about 18

feet in length. One end is cut off fairly near a node and is then

served over with thread, and slightly hollowed. The head consists of a

long piece of iron, such as a stout nail, brought to a sharp point at

one end. The other end is served over with thread in such a way as to

make it fit fairly tightly into the end of the bamboo shaft. Some

distance from the point of the head two barbs of iron are attached by

thread, and between this point and the lower end the line is attached.

(See Fig. 23.)

A man using the harpoon stands on the forward platform of the canoe,

holding the bamboo shaft in his hand. The head is inserted in the upper

end of this, and the line passes over his shoulder and is coiled in the

bottom of the canoe, the other end being attached to the forward

out-rigger boom or to the thwart that takes its place in a large canoe.

He poles the canoe along the reef with the harpoon shaft. When about to

make a throw he raises the shaft till he can hold the butt end in his

right hand, with the point directed towards the fish or turtle, and he

then leaps forward so that if he succeeds in his aim the weapon strikes

with all the force of his weight behind it. When the turtle or fish is

struck the bamboo shaft floats loose and this is secured by the man in

the water, who returns to the canoe. It may be necessary to strike the

prey with a second harpoon, but if the first was well thrown the animal

is firmly held by the line.

Harpoons are not used in the Little Andaman. The natives of the Great

Andaman say that they themselves have only used them since they were

able to obtain iron and that before that time they could only capture

turtle and dugong in nets. It would seem therefore that the harpoon has

been invented or adopted by the tribes of the Great Andaman Division in

comparatively recent times, and was not an element of the primitive

Andamanese culture.

The turtle net is no longer used, as the natives prefer the harpoon and

have all the iron they need. Such nets were formerly made of rope of

Hibiscus fibre. A net was about 150 cm. in width and of variable

length. One specimen that was made for me had an open mesh of about 25

cm. square, while another had a smaller mesh. The knot used in a net

from the South Andaman is the ordinary fisherman’s knot shown in Fig.

24 b. In a net from the North Andaman the knot used is a slip-knot, one

strand being tied with an overhand knot over another which it crosses

at right angles as shown in Fig. 25.

Each end of the turtle net is attached to a stake pointed at the lower

end. The lower edge of the net is weighted with stones attached as

shown in Fig. 24, while to the upper edge are attached a number of

floats, each consisting of a long thin stick of Hibiscus wood to the

upper end of which is attached a tassel of fibre.

The net was placed in shallow water so that the stones rested on the

bottom while the tassels at the upper ends of the floats appeared above

the surface. As soon as a turtle was entangled in the net the agitation

could be observed by those watching who would proceed to the spot to

secure their capture.

So far as is known it would seem that nets of this description are not

used in the Little Andaman.

The Andamanese make practically no use of spears. At the present time

the natives of the Great Andaman Division sometimes make pig-spears of

a length of stout cane or rattan with a head of iron attached. The

natives themselves say that such spears have only been made since the

occupation of the islands, and it is probable that they were first made

in imitation of spears used by Burmese convicts for pig-hunting. They

are hardly ever used, the pig-arrow being preferred.

The true fish-spear is unknown in the Andamans though use is sometimes

made of a harpoon similar to the turtle harpoon, but of smaller

dimensions and with a finer line, for harpooning fish.

In the North Andaman a sort of fish-gig was formerly in use made of

about twelve pieces of Areca wood of about 105 to 110 cm. long and 1

cm. or less broad and with tapered and sharp-pointed ends. These were

fastened side by side by means of a strip of wood near one end, as

shown in Fig. 26. This weapon was used for spearing small fish in pools

on the reef.

CUTTING IMPLEMENTS.

At the present time the Andamanese make use of iron for their cutting

implements. It is uncertain when they first learnt the use of iron, but

it was certainly before the end of the eighteenth century. What iron

they had was obtained from wrecks, of which there have always been a

number on the Andamans. The metal has only become plentiful since the

European settlement of 1858.

It is highly probable that the Andamanese, though they may have learnt

the use of iron from implements used by visitors to their shores, have

not learnt from any other people the method of working the metal. Even

at the present day they do not make any use of heat in the manufacture

of their iron implements, the metal being worked cold. It seems highly

probable that they have simply adopted in connection with iron the

methods they formerly used in dealing with other materials,

particularly shells.

The materials used by the Andamanese, apart from iron, are wood, bone,

shell and stone. We may begin by considering their use of stone.

In former times quartz flakes were used by the Andamanese for the two

purposes of shaving and scarifying the skin, and for hardly any other

purpose. (Among minor uses of stone flakes may be mentioned those of

cutting the finger-nails, and sharpening boars’ tusks.) A quartz pebble

is held in the left hand and is struck with a hard rounded pebble of

any suitable kind. A flake is thus knocked off and falls into the palm

of the left hand. The flake is examined, and if it be suitable it is at

once used. If it be unsuitable it is thrown away and another made. For

shaving, flakes with a sharp blade-like edge are required; for

scarifying, flakes with a fine point are preferred. A flake is used

till its fine edge is lost and is then thrown away and another made.

Thus a woman who is shaving some one’s head may use as many as twenty

flakes one after another, and to obtain twenty suitable flakes she

probably makes as many as forty or even more. The kitchen-middens or

heaps of refuse that are found on the sites of old encampments contain

thousands of quartz pebbles that have been used as cores, and thousands

of flakes.

Besides quartz there is a flinty kind of stone that is used in much the

same way for making flakes. Suitable pieces of the stone are obtained

and are placed in the fire for a few hours. They are taken out, and

when they are cold are used in exactly the same way as a quartz pebble.

At the present time quartz is hardly ever used in this way, for the

natives greatly prefer glass, and they obtain sufficient old bottles

from Port Blair to satisfy their requirements. The bottom of a bottle

is treated in every way as though it was a quartz pebble, a flake being

knocked off and used, and then another and so on till the operation in

hand (whether shaving or scarifying) is completed. The flake is held

between the thumb and first finger when it is being used. In no case is

a flake of quartz or glass ever kept. It is only made when required and

after having been used is thrown away.

The natives themselves say that they formerly never made any use of

stone for cutting purposes save in the case of stone flakes as

described above. As against this there are three statements that must

be considered. Colebrooke, who visited the islands in 1789, says of the

Andamanese that “their canoes are hollowed out of the trunks of trees

by means of fire and instruments of stone, having no iron in use

amongst them, except such utensils as they have procured from the

Europeans and sailors who have lately visited these islands; or from

the wrecks of vessels formerly stranded on their coasts.” The accuracy

of Colebrooke’s statement is made doubtful by the fact that at the

present time (since 1858) the natives do not use fire in making their

canoes, and it seems improbable that if they had this custom in 1789

they should have discontinued it and have entirely forgotten that it

ever existed. Further if they used implements of stone in 1789 it is

certainly strange that by 1858 they should have entirely forgotten that

they ever did so. When Mr Man was making his enquiries the oldest men

all agreed in stating that they never used stone for their adzes. We

may conclude that Colebrooke’s statement is untrustworthy.

Stoliczka records the finding in the South Andaman of a stone celt and

a stone arrow-head in the kitchen-middens of the South Andaman. The

chief reason for doubting the value of this find is that Stoliczka

states that these implements were made of tertiary sandstone, which it

is very hard to believe would be of any use whatever. We may therefore

adopt the opinion of M. Lapicque [254] that Stoliczka had found

fragments of a whet-stone of sandstone and had been mislead into

thinking that he had found an axe and an arrow-head.

A third statement that needs to be considered is one by Mr Portman, who

presented to the British Museum an arrow with a head of stone made

specially for him by a native of the North Andaman who stated that in

former times such arrow-heads were used by the Andamanese. At the

present time the arrow-head is broken and it would seem to be so

fragile as to be entirely worthless for the purpose to which it was

supposed to be put. The natives of the North Andaman whom I questioned

stated that they did not use stone for their arrow-heads, but shell. We

may therefore hold that the evidence given by Mr Portman is not at all

satisfactory.

In the Akar-Bale tribe I heard a legend that at a certain spot there is

a kind of stone which was used by the ancestors for making adzes. I

visited the spot and the stone was pointed out to me. Unfortunately the

specimen that I took was lost and I am therefore unable to state what

the stone was, but it was such that it would have been utterly

impossible to make any sort of adze out of it. It was of a crystalline

nature and was easily fractured by a blow against even a soft substance

such as wood. It was clear that the native statements about it were

merely a legend having no historical value.

We may justifiably conclude that it is probable that the statement of

the natives to the effect that before they possessed iron in any

quantity they made their adzes and arrow-heads out of shell and not out

of stone, is correct. Their use of stone for cutting was therefore

confined to the flakes which have been described.

The most important material to the Andamanese seems to have been shell.

Mollusc shells were used in the natural form or after having been

manufactured. The chief shell used in its natural form is the Cyrena

which serves at the present time as a knife, a scraper and a spoon.

Even when they have knives of iron and steel they still use the Cyrena

shell in preference for some purposes. It is used as a scraper in

preparing fibres for rope and thread, in making arrows, as a knife for

cutting thatching leaves and cane and even thread and rope, and for

making incised patterns on bows and arrows. The shells are always to be

found lying about their encampments, and a few are always carried with

them when they migrate to a fresh camp. Those living inland obtain

their supply of shells from their friends on the coast.

When in use the shell is clasped between the thumb and first finger,

the thumb passing over the convex side and the finger round the hinged

edge. The remaining fingers are used to clasp the object that is being

scraped or cut. In cutting, the motion is away from the body, being

produced by a twist of the wrist. In scraping the motion is away from

the body, or from left to right. A knife of iron or steel is held in

this way by the natives whenever it is used for any purpose for which

they formerly would have used a shell. The blade is clasped near the

handle between the thumb and first finger, the back of the blade

pressing against the root of the thumb, and the handle away from the

body.

Another shell that is used in its natural form is a small whelk shell

that is used as a scraper for scraping off the outer skin of mangrove

seeds in preparing them for food.

A shell that is used in very nearly its natural form is a kind of pearl

shell that grows along creeks through the mangrove swamps. The shell is

only very slightly curved and it is for this reason that it is selected

as suitable. The weak edge or lip of the valve is broken away, and the

edge is then slightly ground on a stone. This implement is used by

women for slicing yams and certain other vegetables such as some kinds

of seeds when they are being prepared for food.

The natives say that before iron was plentiful they used shells for the

heads of their pig-arrows. Several different species of shell seem to

have been used, the chief concern being to obtain a piece of sufficient

size that was as nearly flat as possible. Such shells are those

belonging to the larger bivalves. The natives state that their method

of working the shell was to break it roughly into shape with stones and

then grind it down on a whet-stone until it was given a sufficiently

sharp point and edge. Some arrow-heads of shell were made for me by

this method at my request by one of the old men of the North Andaman.

The natives state that before they had iron they made their adzes of

shell. Two different men of the North Andaman made two shell adzes for

me, one of Pinna shell, and the other of a shell that I omitted to

identify. The Pinna shell adze seemed to me only suitable for light

work such as finishing off a bow or a canoe, as it seemed likely to

break under a strong blow. The other adze was much stronger and

therefore capable of heavy work, and although the edge seemed to me to

make it a poor implement with which to cut down a tree yet it certainly

did not seem less suitable than the stone adzes used by many primitive

peoples. If I were given a choice of implements with which to fell a

tree, between the shell adze of the Andamanese and a stone axe of

South-western Australia I should certainly choose the former.

According to the natives they formerly used bone for the points of

their fish-arrows and for the barbs of their pig-arrows. For both of

these purposes they now use iron. The bone was broken into a piece of

suitable length and then ground down with a whet-stone. Apparently the

bone most frequently used was the tibia of the pig. For their

fish-arrows they also made use of the bone of the tail of the

sting-ray,—its “sting.” When the fish was caught the bone was knocked

off and reserved for use. It required no treatment whatever, being

simply bound on to the point of the fore-shaft in the same way that an

iron point is now attached.

Colebrooke, in 1789, described their arrows as “headed with fish-bones

or the tusks of wild hogs, sometimes merely with a sharp bit of wood

hardened in the fire.” By fish-bone he probably means the bone of the

sting-ray. Where he writes the tusks of wild hogs we should probably

read “the bones.” A boar’s tusk is curved, and it seems impossible to

imagine how it could possibly be used as an arrow-head or arrow-point.

The boar’s tusk is used by the Andamanese as an implement, however,

making a very efficient sort of spokeshave. The edge which is used is

kept sharp by scraping with a quartz or glass flake or with a Cyrena

shell. The edge is near the point (at a in Fig. 27) and the tusk is

clasped at the other end between the forefinger and the root of the

thumb (at b in Fig. 27), the movement being away from the body. It is

used for planing bows and paddles, and in the hands of an Andaman

Islander is a very efficient implement, producing a beautifully smooth

and even surface.

Of wood the Andamanese formerly made knives and arrow-points. The

knives were made of a slip of bamboo or cane shaped and sharpened with

a Cyrena shell. Such knives were used for cutting meat and apparently

for no other purpose. A knife was always attached by a short length of

cord to a skewer of pointed Areca wood. The double implement was used

in cooking and eating, the skewer serving to lift pieces of meat in and

out of a pot, while the knife served to cut them. At the present time

the cane or bamboo knife is replaced by a knife made from hoop-iron,

but the shape of the original implement is retained as nearly as

possible and a skewer of Areca wood is generally attached to it.

As has been already stated the Andamanese formerly used hard wood such

as that of the Areca palm for the points and heads of their arrows.

They do not seem to have made use of bamboo in this way.

At the present time iron is used for the blades of adzes, for the heads

and barbs of pig-arrows, the points of fish-arrows, the heads of

harpoons, and for knives. The method of working the metal is apparently

exactly the same as the method they formerly used for working shell and

bone. For the head of a pig-arrow a suitable piece of iron is taken and

a fragment of about the right size is broken off by means of a stone

hammer. This is then roughly broken into the required shape, no heat

being used, and no advantage being taken of the malleability of the

metal. The next process is to grind it on a whet-stone. The natives are

always eager to obtain files which enable them to do this part of the

work much more rapidly. When the arrow-head has been ground into shape

the edges are sharpened. Blades for adzes are made in exactly the same

way from any suitable piece of iron or steel, such as a cutlass or an

old file or a piece of thick hoop-iron. The adze-blade is attached to a

handle of mangrove wood by the method shown in Fig. 28.

The barbs of pig-arrows and the points of fish-arrows are made in the

same way by breaking the metal into a suitable shape and then grinding

it on a whet-stone. It is probable that this was the method that was

formerly used for dealing with bone for these purposes.

In the case of the knife, iron or steel is now substituted for cane or

bamboo, but the knife has retained its shape in spite of the change of

material. The shape of a knife, whether of cane or of iron, with its

attached skewer of Areca wood is shown in Fig. 28.

In the case of the harpoon the native tradition is that this implement

was only made after they had discovered the use of iron.

At the present time both the Semang and the Negritos of the Philippines

make use of iron. The Semang heat the iron until it is red-hot and then

batter it into shape between two stones [255]. The shapes of the iron

weapons and implements which they make follow fairly closely those made

by the Malays.

In an attempt to reconstruct the primitive Negrito culture it would

seem that the most reasonable hypothesis is that the primitive Negritos

had no knowledge of iron and had not learnt to fashion implements out

of stone, but relied entirely on such materials as wood, bone and

shell. The Andamanese, becoming possessed of iron through wrecks upon

their islands, applied to it the technique that they had developed for

dealing with shell, and thereby invented their present method of

working iron without heating it. The Semang and the Negritos of the

Philippines probably first learnt the use of iron from their neighbours

of other races. There is not at present any evidence to show that the

Negritos ever had any method of working stone except the very simple

one at present in use in the Andamans for making flakes.

STRING, ROPE, MATS, BASKETS, AND NETTING.

For string, rope and thread the Andaman Islanders make use of a number

of different vegetable substances, but they make no use whatever of any

animal substances. Some of the more important fibres, with their uses

and the methods of preparing them, are mentioned below.

The bark of the Hibiscus tiliaceus, which occurs in the beach

vegetation in all parts of the islands, provides the Andamanese with

one of their most important fibres. By the coast-dwellers of the Great

and Little Andaman Divisions it is used for making rope. In the Great

Andaman Division the rope made from it was formerly used for making

turtle nets, and is now used for the lines of turtle harpoons, and for

hawsers to attach a canoe either to a stone used as an anchor or to a

tree. No other fibre is used for these purposes. The Hibiscus rope does

not seem to be much affected by salt water. The forest-dwellers of the

Great Andaman have less use for rope, and at the same time are not able

to obtain so readily the Hibiscus fibre. What rope they do have is

therefore obtained from the coast-dwellers or is made of some other

fibre. In the Little Andaman the Hibiscus is regularly used for rope.

It is also used for the short cord by which the detachable head of the

pig-arrow is attached to the shaft. In the Great Andaman a strip of the

bark of the Hibiscus is used for the sling in which children are

carried. Strips of the bark are worn by the women of the Little Andaman

across their shoulders and breasts, as a sort of ornament.

To obtain the fibre young straight shoots of about 120 cm. in length

are cut from the tree, those free from gnarls, and having a smooth

bark, being chosen. The bark (inner and outer layers) is peeled off in

strips of from 1·5 to 3 cm. in width. The inner or liber layer is then

separated from the outer layer of the bark, is well scraped with a

Cyrena shell and dried in the sun or over a fire. When dry it is worked

in the hands until the various layers of fibre separate one from

another. It is then ready to be made into rope. The fibre is

interlacing, and when freshly made is a lustrous greyish brown. After

exposure to salt water it turns a dark brown.

Mr E. H. Man, in his work on the Andamans, speaks of the Melochia

velutina as providing fibre for rope. This is an error. The tree to

which Mr Man refers is the Hibiscus tiliaceus. It is extremely common

on the shores of the islands, as it is in many other parts of the

tropics. It is very easily identified, as it bears its characteristic

yellow flowers for a long time every year. There is no doubt whatever

that this is the tree from which the natives regularly obtain the fibre

for the rope they use in turtle-hunting and fishing and in their

canoes. Other writers, following Mr Man, have repeated his error in

calling it the Melochia velutina, for example Sir Richard Temple and Mr

Portman. In Mr Portman’s collection of photographs in the British

Museum there is a good photograph of a Hibiscus tiliaceus tree labelled

“Melochia velutina.” I looked carefully in the Andaman jungles for the

Melochia velutina but was unable to find it, and I am quite certain

that in any case, even if it be found there, it is not commonly used by

the natives for rope.

The bark of a number of other trees provides fibre of which the natives

occasionally make use. Amongst these are one or more species of

Sterculia (S. villosa?), and a tree that I identified somewhat

doubtfully as Grewia laevigata. The coarse fibre of the liber layer of

these trees may be made into rope by the same method as that employed

in dealing with Hibiscus tiliaceus. Very little actual use is made of

them however. A fibre which looks very like that obtained from one

species of Sterculia is frequently used by the natives of the Little

Andaman (and also by the Jarawa) for their personal ornaments. Mr

Portman says that this fibre is obtained from the Celtis cinnamonea.

A species of Hibiscus, which I believe is Hibiscus scandens, growing in

the jungles and not along the shore, provides a fibre that is prized by

the natives of the Great Andaman tribes for making string or fine rope.

The fibre is less easily obtained than that of the Hibiscus tiliaceus,

but owing to its quality (it is not so interlacing) is capable of being

made into finer rope and string. It is often made into string and then

used for making netted bags. I did not find this fibre in use in the

Little Andaman, but as it is not very often met with even in the Great

Andaman, it may possibly be used in the Little Andaman.

There are several species of Ficus in the Andaman forests, and the

natives know that they can obtain fibre from the bark of these trees.

The only one that is regularly made use of is the Ficus laccifera. The

natives of the Great Andaman Division use the bark of this tree for

making their personal ornaments. In the Little Andaman it is used for

bow-strings. A fibre called in the Little Andaman ulu, and said by Mr

Portman to be obtained from the Ficus hispida, is used in that island,

and by the Jarawa, for making personal ornaments.

The Gnetum edule, a climbing plant that is fairly common, is used in

all parts of the Andamans for thread and string. The creeper is cut

into short lengths at the nodes and is dried for a few days. The outer

layer of bark is then scraped off with a Cyrena shell, and the liber

layer beneath it is peeled off in fine strips and these are made into

thread or string. This string is used in the Great Andaman Division for

making netted bags and fishing nets. In the Little Andaman it is used

for binding their arrows, as well as for netting.

The most valuable fibre of the Great Andaman tribes is that of the

Anadendron paniculatum, which is used for string and thread and for

bow-strings. Until recent times the method of preparing the fibre was

not known to the natives of the Little Andaman, but they have now

learnt it from the natives of the Great Andaman with whom they have

been brought in contact, and the use of the fibre for string and thread

is coming in amongst them.

The fibre is not easy to prepare. Long thin branches of the creeper are

cut, which must be neither too young nor too old. To obtain these it is

often necessary to climb up into high trees, for the Anadendron is a

climbing plant. The creeper is cut into lengths of from 20 to 40 cm.

The bark (inner and outer layers) is peeled off these in strips of from

7 to 10 mm. in width. A strip of the bark is taken and placed on the

thigh, inner surface downwards, and is scraped with a Cyrena shell

until the outer bark is entirely removed and the fibres remain clean

and separate. These are dried in the sun or over a fire and, if not

needed for immediate use, are stored for future occasion. The fibre is

fine and of a light greyish brown colour. In its qualities it somewhat

resembles ramie fibre. It is extremely strong.

There are a number of other trees and plants that are known by the

natives to afford fibre, but they are not used, or if they are, it is

extremely rarely.

The Andaman jungles have a number of different species of Calamus, and

the canes or rattans of these are put to all sorts of uses, such as the

making of baskets, the lashings and furnishings of canoes, and in

building huts. In the Little Andaman one species of cane is used to

provide the fibre for women’s belts. The outer skin is removed and the

remainder of the cane is divided into fine strips or threads. A bundle

of these tied together constitutes the belt worn by the women of the

Little Andaman. The outer sheath of the leaf-stem of the Calamus

tigrinus is used by the natives of the Great Andaman Division for

making mats. Lengths of the leaf-stem are cut and the outer skin is

removed in strips of about 3 to 5 mm. in width. The still adhering pith

is removed with a Cyrena shell, and the strips are dried in the sun and

then made into mats.

The leaf-stem of a species of palm is cut while green and is then

shredded into long strips. The fibre thus obtained really consists of

the leaflets of the young unopened leaves of the palm. It is used in

the Little Andaman to make the tassel that women wear over the pudenda.

The women of the North Andaman formerly wore a tassel of this fibre,

but have now discontinued the custom, since their contact with the

South Andaman. In the Great Andaman tribes this material has important

ceremonial uses. It is called ko̱ro in Aka-Jeru and ara in Aka-Bea, and

has been frequently mentioned in this volume. A tassel of the fibre is

suspended near the grave of a dead person and at the entrance of the

village at which the death took place. In the North Andaman a suspended

cane hung with a fringe of the fibre is erected for the peace-making

ceremony. (See Plate XIX.)

Two plants that were not identified are used in the North Andaman for

making baskets. The methods of preparing these will be described later.

The natives of the Great Andaman Division make use of the leaves of the

Pandanus Andamanensium for making belts for women and ornaments that

are worn on ceremonial occasions. These leaves do not seem to be used

in the Little Andaman.

The pods of one or more species of Dendrobium are collected by the

natives of both Great Andaman and Little Andaman. They are roasted in

the fire, until the outer skin turns a bright yellow, and this is torn

off in strips and used for ornamenting nets, baskets, rope, etc.

The above description includes all the more important vegetable

substances used by the Andamanese for their rope, string, netting, and

basket-work. There are many other substances that they might use if

they wished, of the properties of which they are fully aware. Their

knowledge of the trees and plants of the forests and of the peculiar

properties of each is very extensive. They themselves say that they use

only those that best serve their purposes.

The Andamanese make rope and string or thread, but in all cases it is

only two-ply. Rope is made by men only, and is used for the lines of

turtle harpoons, and was formerly used for turtle nets. The ropes made

from Hibiscus fibre are very strong and durable, being quite as good as

the best hempen ropes of the same diameter. In rope-making the Hibiscus

or other fibre (Sterculia or Grewia) is taken and twisted into a long

strand, either with the fingers, or on the thigh by rolling beneath the

palm of the hand, short lengths of fibre being added until a single

twisted strand of sufficient length and uniform thickness is produced.

The middle of this strand is passed over a piece of wood held by the

toes, one half of it being wound on to a reel (kutobi in Aka-Jeru) made

by tying together crossways two pieces of cane or wood each about 20

cm. long and 6 mm. in diameter. The other half of the strand is loose,

and is held (near the point where it is tied to the wood held in the

toes) between the finger and thumb of the left hand, the rest of it

passing across the palm, over the left forearm, under the armpit,

across the back and over the right shoulder, hanging down loosely to

the worker’s right side. This arrangement is in order that the loose

strand shall not become entangled or get in the man’s way as he works.

The reel is held in the right hand and is passed first under the left

hand, then back again over it, the two strands being thus twisted into

a firm two-ply cord.

The natives of the Little Andaman make rope in much the same way, but

they pass the reel from right to left over the other strand and back

under it, the twist being thus in the opposite direction from that used

in the Great Andaman.

String or thread is made by both men and women. It is put to many uses,

the chief being for binding the heads of arrows, harpoons and spears

and the ends of bows, and in making nets, baskets, mats and personal

ornaments. In making string the man or woman sits down with legs

outstretched. Thin strands of fibre, varying in thickness according to

the thickness of the string required, are taken and each twisted singly

by being rolled between the palm of the right hand and the right thigh,

the motion being away from the body. When a sufficient number of short

single strands has been thus made, two of them are taken and placed

together on the thigh, being held at one end in the left hand. The two

strands are rolled together beneath the palm of the right hand, the

motion being inwards towards the body. A well twisted thread is thus

produced. When some 10 cm. or so have been thus twisted, the thread is

rolled once beneath the palm of the hand in the opposite direction,

i.e. away from the body, this action rendering it more compact. As soon

as the end of the two strands that are being twisted is neared, two

more are taken and joined on, first one and then the other, by being

rolled in with the first two. Fresh strands are thus continually added

as the string grows in length. String of any desired length is made in

this way, of considerable strength and of surprisingly uniform

thickness.

String is made in this way from the fibre of Anadendron, Gnetum and

Hibiscus scandens in the Great Andaman Division, and from Gnetum and

ulu fibre in the Little Andaman. In the Great Andaman string made from

Anadendron fibre is rendered more durable by being waxed with black

bee’s-wax, but this treatment is not considered necessary for string

made from Gnetum fibre or from Hibiscus scandens.

Ornamental rope is made for men’s belts in the Great Andaman. Hibiscus

fibre is twisted into a single strand. Around this strand strips of

Dendrobium skin are wound spirally so that it is entirely covered, and

the strand itself is twisted into a two-ply cord.

Two other forms of cord have been already mentioned, namely the

bow-string, of twisted fibre of the Ficus laccifera in the Little

Andaman, and of wrapped fibre of the Anadendron in the Great Andaman

Division, and the special cord used in the Great Andaman for attaching

the head of a pig-arrow to the shaft. The Andamanese make very little

use of plaited cord. I have only met with it in personal ornaments made

of Pandanus leaf in the Great Andaman.

The mat-work of the Andamanese is very simple. The natives of the

Little Andaman make bamboo mats on which they sleep. Strips of bamboo

of about 120 cm. in length and ·75 cm. in width are attached by means

of thin strips of cane to other strips of bamboo at right angles to the

first series. The technique is illustrated in Fig. 29 which shows the

back and the front of a portion of such a mat. It is that usually known

as wrapped-twined work.

A similar technique is used in both the Little Andaman and the Great

Andaman in making thatching of cane-leaves. There is a difference,

however, the wrapping used in making thatch being that shown

diagrammatically in Fig. 30.

The natives of the Great Andaman make sleeping mats from the outer

sheath of the leaf-stem of the Calamus tigrinus. Lengths of the

material are prepared and cleaned and are cut to a uniform length,

generally about 60 to 80 cm., having a breadth of 3 to 5 mm. A length

of thread, generally of the less valuable Gnetum fibre, but

occasionally of Anadendron fibre, is made and is wound on to two

netting needles, one half on each. With this thread the strips of

cane-leaf are fastened together. The technique is different in the

North Andaman and in the South Andaman.

In the North the technique, which is represented diagrammatically in

Fig. 31, A is that known as wrapped-twined work. One of the two threads

of which the work is composed is held taut, the needle on which it is

wound being held in the toes, and the other thread is wrapped spirally

round it, one of the strips of cane-leaf sheath being enclosed at every

turn. Thus the method is exactly the same as that adopted in making

mats of thatch.

In the South the technique is that known as twined work or fitching.

The two wefts (i.e. the threads) are twisted together in the same

direction one under the other, enclosing at each half turn one of the

strips of leaf-sheath. This is shown diagrammatically in Fig. 31, B.

The mat is made by parallel lines of such twining or wrapped-twining.

At each side of the mat the line of threading is quite close to the

ends of the strips of which the mat is made. In the South Andaman the

work on each side of the mat is different from that in the middle. Each

of the two threads is alternately given a complete turn round one of

the strips of material.

Mats may be of any length, and examples vary from 1 metre to 10 metres.

When in use about a metre and a half is unrolled and the remainder

serves as a pillow. If the mat be short a split log does service as a

pillow instead, but the full-length mat is certainly a more comfortable

bed. When a mat begins to wear or fall to pieces in some part it is not

thrown away, but this part is kept rolled up, and an unbroken part of

the mat is unrolled to sleep on, the remainder being kept rolled up

either at the head, where it forms a pillow, or at the foot. An old mat

may be continually increased in length by additions made to it. The

work of making these mats is performed by women only.

The Jarawa and the natives of the Little Andaman make mats of a

somewhat similar pattern, but I have not been able to secure one so as

to see how it is made.

To explain the different forms of the Andamanese baskets, it is most

convenient to begin by considering the way in which the natives of the

North Andaman tie up their pots. The small cooking-pots of the North

Andaman are fragile things and are not easily made, and it is therefore

necessary to take care of them. A leaf of the Licuala palm is taken.

The leaflets are plaited over one another, close to the point of their

insertion in the petiole, so as to form a sort of rosette with leaflets

radiating from it in every direction (Fig. 32 a). This is laid on the

ground, the pot is placed on it upside down, and the leaflets are

brought up all round the pot so as to meet at the point, and are there

roughly fastened. Three strips of cane are then taken and are tied

together crossways in the middle so as to form a sort of six-rayed star

with six approximately equal angles. This is laid on the ground with

the outer surface of the canes downwards, the pot is placed on it

upside down, and the strips of cane are bent upwards over the pot so as

to meet at the point. The ends of two opposite strips of cane are left

projecting for a few centimetres above the point, and the ends of the

other four are fastened firmly down. Another strip of cane is now

fastened round the middle of the pot, being applied to the six canes

previously mentioned by wrapping, i.e. a turn is taken with it round

each of the six strips in turn. The pot is now safely tied up and can

be hung in the hut or carried on a journey without much fear of its

coming to grief. To use the pot it is of course necessary to untie it.

A pot wrapped up in this way is shown in Fig. 32 a and b.

In the South Andaman the pots have rounded bottoms. They are not tied

up in the same way as in the North Andaman, but for purposes of safety

and carriage each pot is provided with a rude basket. The basket is

made so as to fit the pot. Six strips of cane are taken and tied

together in the middle, as previously described. A stout strip of cane

is then taken and is bent round into a circle so as to be just a little

larger than the outside rim of the pot. This, which is to form the rim

of the basket, is placed in position and the six strips of cane before

mentioned are bent round and attached to it. The manner of attaching

the uprights of the basket to the rim is shown in Fig. 33. The strip of

cane forming the upright is thinned down beyond the point where it

reaches the rim. It is given one turn round the rim, ascending on the

outside and descending inside, then a half turn round the standing part

of itself, immediately below the rim from left to right, then another

complete turn round the rim ascending inside and descending outside, a

half turn behind its own standing part below the rim, a third complete

turn round the rim, and the end is fastened with an overhand knot.

A thinner strip of cane is now taken and wound round the six uprights

(the warp of the basket) being given a turn round one after another.

The technique is that known as wrapped work. If the strip be not long

enough another is joined to it with a reef or sailor’s knot. The weft

(as this thinner strip may be called) is given five or six spiral

turns, and thus leaves a very open and rather weak basket. The basket

is then further strengthened by other strips of cane attached by one

end to the rim and carried downwards, with one turn round each of the

horizontal canes (weft) and the other end attached to the centre of the

bottom of the basket. A strip of cane or bark is attached to the rim by

its two ends to provide a strap by which the basket with its pot may be

carried on the back. Before the pot is placed in it the basket is lined

with the leaflets of the Licuala palm.

We may now turn to the baskets of the Little Andaman, of which there

are two varieties, one made with more care than the other. As regards

shape both varieties are the same, and the shape is exactly that of the

Little Andaman pots.

The following is a description of a small but typical specimen of the

better variety of Little Andaman basket. The foundation consists of

twelve whole canes. A little under 80 cm. of the cane is left whole,

and at each end it is thinned down to a strip. The twelve canes are

taken in four bundles of three each and placed so as to cross each

other in the middle. They are bent into a somewhat conical shape, and

the ends (where the cane is thinned) are attached to a rim composed of

a whole cane bent into a circle and tied. The method by which each of

the uprights or stakes is attached to the rim is shown in Fig. 34. It

is almost identical with that used in the South Andaman pot basket. The

weaving is then begun near the bottom of the basket. A thin strip of

the outside of a cane is taken, and is applied to the uprights (warp or

stakes) by wrapped work, i.e. it is given a complete turn round each of

them in turn. This wrapped weaving is continued spirally from near the

bottom of the basket to near the rim. Near the top of the basket,

between 4 and 5 cm. from the rim, the weaving is so arranged that for

about three-quarters of the way round the basket there is a gap of

about 1·5 cm. between two rows of weft (see Fig. 34). The purpose of

this will be mentioned later. The basket now consists of a rim to which

are attached twelve uprights forming the warp of the basket, around

which a fine strip of cane has been spirally wrapped from the bottom to

the top. Twelve fine strips of the outside of a cane are now taken. One

end of each is fastened to the rim in between two of the stakes, the

mode of fastening being shown in Fig. 34. The strip is then carried

down the basket, on the outside, as far as the bottom row of weaving,

round which it is doubled, and is then wound spirally, from the bottom

to the top of the basket, around its own standing part, including at

each spiral turn one of the wefts. Thus each of these twelve strips of

cane is attached to the outside of the basket by the process known as

wrapped-twined work. The basket is furnished with a handle of bark

fibre which is attached by its two ends to the rim on that side where

there is no gap between the rows of weft near the top.

In the less carefully made baskets of the Little Andaman there are a

few important differences. The stakes of the basket are not thin whole

canes, but are strips of larger canes. The weft is applied in wrapped

work, as in the basket already described, but the rows of weft are not

so close together and are therefore not so numerous. After the first

process, when the basket consists of stakes and horizontal (spiral)

weft only, vertical strengthening strips are added, but these are

applied, not in wrapped-twined work as in the basket described, but in

wrapped work, in exactly the same way as in the South Andaman

pot-basket. Thus, apart from its shape, which is that of the Little

Andaman pot, the Little Andaman basket of this kind is very similar to

the South Andaman pot-basket.

The natives of the Little Andaman make pots that are much larger and

deeper than those of the South Andaman, and have a more rounded bottom

than those of the North Andaman. For every pot a basket is made that

exactly fits it, and in this basket it is stored and carried. Every

basket that is made in the Little Andaman, whether it be used for

carrying a pot or for any other purpose, is made of exactly the same

shape.

The purpose of the gap that is left near the top of the basket as

described above is in order that strips of string or fibre may be tied

across the mouth of the basket, from side to side, in order to keep its

contents safe.

We may now return to the Great Andaman and consider the baskets of the

South Andaman Group of tribes. These are made from the best canes. From

80 to 120 fine strips of cane are taken which are to form the stakes or

uprights. A slight hollow is made in the ground, and the strips of cane

are placed crossways across one another in this hollow, the inner

surfaces of the canes being downwards. As the strips are being

arranged, and when the weaving is begun, the centre, i.e., the point

where the strips cross one another in the middle of the hollow, is

pressed firmly beneath the heel so as to maintain them in position. The

first few strips are sometimes tied together in the middle, but this is

not always done. When all the strips are arranged evenly the weaving of

the basket is begun with a length of thread, which is slewed in and out

between the strips of cane, beginning as near the centre as possible,

the stroke being that of ordinary wicker-work. After four or five

spiral turns have been taken with the thread it is fastened. The bottom

of the basket is then reversed, the stakes being bent over, and the

weaving proper is begun with a fine strip of cane. This is applied by

wicker-work nearly as far as the top of the basket. At the top, the

weaving is finished off with three or four spiral turns of

wrapped-twined work in cane (see Fig. 35). The rim of the basket is

formed of a thin piece of wood (circular in section) bent round into a

circle of the right size and the two ends tied together. The stakes or

uprights are attached to this rim (after the weaving is finished) but a

space of about 5 cm. is left between the top of the weaving and the

rim. The mode of attaching the uprights to the rim is shown in Fig. 35.

The South Andaman basket is really a conical basket with the bottom

reversed or dented inwards to form a “kick” like the kick of a bottle.

If it were not for the kick it would be the same shape as the Little

Andaman basket. The kick enables it to stand upright, although it is

inclined to be top-heavy, but renders it unfit for carrying pots.

The space between the top of the weaving and the rim is to admit of

strings being tied across the mouth of the basket to keep its contents

safe.

A handle of Hibiscus fibre is attached to the rim and rests across the

front of the chest when the basket is carried on the back.

South Andaman baskets are sometimes ornamented, in the process of

making, with strips of Dendrobium skin, applied horizontally by

overlaid interlacing. The strip of Dendrobium skin is laid over the

weft and woven in with it for one turn round the basket. I have never

seen ornamentation with Dendrobium skin applied to the South Andaman

baskets by any method save this one.

Patterns of red paint and white clay are occasionally painted on

baskets when they are newly made. Shells are sometimes attached to

different parts of the basket by thread, for the purpose of ornament.

In the North Andaman, baskets are made that differ in several important

features from those of the South Andaman. They are not made of cane but

of two different materials. One of these is the stem of a creeper

called čup-to̱i. Lengths of the creeper are cut and dried and then split

lengthways into two or three pieces according to their size. The outer

bark or skin is then scraped off with a Cyrena shell. These strips are

to form the stakes or warp of the basket. The other material is another

creeper called bobi. The long tough tendrils of this plant are taken

and the soft outer sheath removed by drawing the tendril through a

piece of split cane or bamboo bent double so that the tendril is

scraped between the two inner surfaces. The fibre that remains is split

longitudinally into two pieces and dried.

A bundle of strips of the čup-to̱i is taken, sufficient in length and

number for the required size of basket. The bundle is divided into two

equal bundles and these are tied together in the middle crossways with

thread. This cross forms the beginning of the basket. The weaving is

begun with thread, which is slewed in and out between the warp, from 4

to 7 spiral turns being made. During this process the centre of the

basket, i.e., the cross, is pressed beneath the heel into a slight

hollow made in the ground, to give it a curve which, in the finished

basket, will form the “kick.” The thread is tied, the bottom is turned

upside down, the stakes are bent back and the weaving is continued, not

with thread but with strips of bobi, till a short distance from the top

of the basket, and then three or four rows of wrapped-twined work are

made, the strip of bobi being coiled round the basket and attached to

the uprights with thread. A rim is made of a strip of cane bent into a

circle. This is placed inside the uprights and tied to them in two or

three places. The remaining portion of each of the uprights, projecting

above the rim, is bent down outside and slightly obliquely, and tied

down by a thread passing over each in turn and round the rim. A very

rough and untidy rim is thus produced, and this is again served or

bound over with thread. A handle of Hibiscus fibre is added.

In the North Andaman baskets, as in those of the South Andaman, a space

of a few centimetres is left between the top of the weaving and the

rim, there being for that space only warp or uprights and no weft. This

allows string to be tied in any direction across the mouth of the

basket, so as to keep its contents safe.

The shape of the North Andaman baskets is different from that of the

southern baskets, the former having a sort of belly at the bottom and

narrowing somewhat above. The result is that the northern baskets will

stand more firmly, being less top-heavy when either full or empty than

those of the south.

In the North Andaman baskets are ornamented as they are made with

worked-in strips of Dendrobium fibre. There are several different

methods of working this ornamentation, resulting in different patterns.

Baskets are also ornamented, when new, with painted patterns in white

and red, though this is not general, and occasionally shells are

attached to them by thread.

One more form of basket-work remains to be briefly mentioned. The

natives of all parts of the islands were formerly in the habit of

preserving as trophies the skulls of pigs and turtle that were killed

in the chase. The natives of the Great Andaman Division do not now

trouble to preserve all the skulls of the pigs they kill, and they give

as their reason for this that now that they have dogs the hunting of

pigs is not a sport that requires any great skill. The Jarawa however

still keep up the old custom, and they go so far as carefully to encase

every skull in basket-work. As may be seen from Fig. 36 the basket-work

in question is of simple wrapped work, the material being strips of

cane.

It is of some interest to consider the different forms of technique

used by the Andamanese in dealing with flexible materials. Rope and

string are only made two-ply. It would seem that the Andamanese have

not discovered that three-ply cord is stronger for a given diameter

than two-ply. They have, in the bow-string of the Great Andaman, an

interesting form of cord that may perhaps best be described as wrapped

cord. The making of a rope involves the twisting of two strands of

material around one another. The making of a wrapped cord involves the

spiral wrapping of one strand of material round another. This is

exactly the same process as “serving,” and it is one that is used by

the Andamanese in all sorts of ways. In serving their arrows with

thread and in serving the ends of bows and the heads of harpoons the

Andamanese have several different methods of making fast the ends, and

I regret that I did not take more detailed notes on this subject. Their

skill in handling this technique is shown in the strength of the

binding on their arrows.

This spiral wrapping of one strand round a flexible or rigid object

lies at the base of much else in their technique. We have seen that

they make considerable use of wrapped-twined work. In this work a strip

of material crosses at right angles a number of strips of the same or

other material, and a weft is wound round the former, taking in one of

the latter at each turn. There are two methods of doing this, either by

simple spiral wrapping, as in the mats of thatch, or by what may be

called “right and left” or “zig-zag” wrapping, as in the bamboo mat of

the Little Andaman. A difference is also made according as the strip of

material around which the weft is wrapped is rigid, as in the Little

Andaman bamboo mat or in thatching, or flexible, as in the North

Andaman mat, where it is one thread while the weft is another thread.

Wrapped work, in which a strip of weft is wound successively round one

after another of a number of rigid stakes, is another very simple

process that is employed in a number of different ways by the

Andamanese. The most important development of wrapped work amongst them

is seen in the pot-basket of the South Andaman, in the baskets of the

Jarawa and Little Andaman and in the pig’s-skull basket-work of the

Jarawa.

Simple twined work is rare in the Andamans. There is hardly any example

of it except in the mats of the South Andaman. It would seem probable

that the North Andaman mat technique of wrapped-twined work is the

earlier, being more in agreement with what we may call the technical

habits of the Andamanese, and that the South Andaman mat technique is a

later elaboration. In this connection it may be remembered that

plaiting, in which also several wefts are twisted one over another, is

rarely used in the Andamans.

The process of nippering, by which the natives of the Great Andaman

make the cord of their pig-arrows, and the somewhat similar process

used in the Little Andaman in making personal ornaments, are quite in

accordance with the general trend of the technique, but when such a

process is applied to a number of parallel strips of material instead

of to two only it constitutes a step towards wicker-work. It is

notable, however, that it is only in the Great Andaman that wicker-work

is used, and this suggests that it has only been invented or adopted

since the separation of the two divisions of the Andamanese.

It is very tempting to regard the different forms of basket, in the

order in which they are described above, as so many stages of a process

of evolution. It is, at any rate, worth while to state the argument,

and to show what the differences between them exactly are. In the North

Andaman pot-covering we have (1) the technique simple wrapped work, and

(2) the basis six strips of cane tied together in the middle. In the

South Andaman pot-basket we have both these features, but the

difference in the shape of the pots allows them to be carried in a true

basket and we have therefore (3) a rim, with (4) a peculiar method of

attaching the uprights to the rim, (5) a number of horizontal (wrapped)

wefts instead of one or two, and (6) strengthening strips applied to

the horizontal wefts in wrapped work. In the rougher kind of basket

made in the Little Andaman we have nearly all the features of the South

Andaman pot-basket. The only differences, apart from the shape of the

basket, which in each case follows the shape of the pot, are (2) more

than six uprights may be used, and (5) the number of horizontal wefts,

i.e. the number of spiral turns taken round the basket, is as a rule

greater. In the more carefully made Little Andaman basket there are

several differences. The uprights are fine whole canes instead of

strips of split cane. It is undoubtedly more difficult to procure whole

canes of the proper size than simply to split up larger canes, but

where the technique is wrapped work the circular section of the

uprights improves the quality of the resulting basket, as a strip of

cane is more easily wound round a whole cane than round a split cane,

and there is less chance of it breaking when a strain is put upon it.

Another difference is that in the better Little Andaman baskets the

rows of weft are as close together as the peculiar technique will

allow. This makes a finer and stronger basket, and is an obvious

improvement. The third difference is that in these baskets the

strengthening strips are applied not in wrapped work, but in

wrapped-twined work, which, however, we have seen is a common technique

in the Andamans. In the better baskets of the Little Andaman we find a

special feature of some interest in the gap that is left in the weaving

near the rim, to allow of strings being tied across the mouth.

When we look at the South Andaman basket the first thing that strikes

us is that it is really a conical basket of much the same shape as the

Little Andaman basket, but with the bottom dented in to make a kick, so

allowing the basket to stand on its bottom. This denting is only

rendered possible, however, by the fact that the uprights of the South

Andaman basket are thin strips of cane that can be easily bent, and

this again depends on the use of wicker-work in the basket instead of

wrapped work. It must be remembered that the top of the weaving is

finished off with three rows of wrapped-twined work, and this suggests

that there may possibly have been a stage of development between the

Little Andaman basket and the South Andaman form, in which the uprights

were thin strips of cane, and the weft was applied in wrapped-twined

work from top to bottom. This, however, is only a surmise. It does

certainly seem probable that the South Andaman basket is derived

immediately from a form of basket similar to that of the Little

Andaman, the great difference being the change to wicker-work

technique. The method of attaching the uprights to the rim was

doubtless introduced owing to the fact that the original method is

unsatisfactory when the uprights are thin and easily broken strips,

instead of stout ones.

The North Andaman basket seems to have been derived from one similar to

that of the South Andaman by the introduction of two changes, (1) the

use of different materials, and (2) the change of shape. The materials

used in the north are such as to give a basket on the whole stronger

and more durable than that of the south. If a heavy weight be carried

in a southern basket the pressure of the basket on the back tends to

crack the canes of which it is composed. In the case of a northern

basket it may lose its shape, but the materials of which it is composed

will give or bend without cracking so readily. The shape of the

northern basket is certainly an improvement, as it avoids the

top-heaviness of the southern shape. Both the northern and the southern

baskets have a gap between the top of the weaving and the rim, like the

basket of the Little Andaman.

Thus every step, or nearly every step, in a hypothetical process of

evolution is exhibited in the different forms of basket-work. First we

have the pot-covering of the North Andaman, then the pot-basket of the

South Andaman, then the rougher kind of Little Andaman basket, of the

same shape as the pot, then the South Andaman basket of cane with a

kick and finally the basket of the North Andaman. The better kind of

Little Andaman basket is simply an independent improvement of the

other, involving no new technique.

However much or little probability we may attach to this hypothetical

reconstruction of the history of basket-work in the Andamans, one thing

does seem fairly certain, and that is that the original ancestors of

the Andamanese were not acquainted with wicker-work, or had no use for

it. In the Little Andaman Division only wrapped work and wrapped-twined

work are used, and the wicker-work of the Great Andaman Division has

almost certainly been adopted since the two divisions were separated.

The consideration of the general technical bias of the Andamanese in

their dealings with flexible materials supports the view that in their

case wicker-work is later than wrapped work and wrapped-twined work. It

seems more than likely that the Andamanese of the larger island have

invented wicker-work in its simplest form on the basis of a previous

technique of wrapped and wrapped-twined work. To us wicker-work seems

such a simple process as almost to need no inventing. It must be

recognized however that the general bias of the Andamanese is against

using materials in this way. The Andaman Islander shows a decided

preference for those processes in which he uses a single flexible

material which he winds or wraps round other rigid or flexible

material, as in nippering, or wrapped work or wrapped-twined work.

It is impossible to obtain confirmation of this view, however, from a

comparison of the Andamanese with the Semang and the Philippine

Negritos. The Semang make mat-work bags and wallets of check, and they

make (or use) baskets of hexagonal work. Both check and hexagonal work

are used by other races in the Malay Peninsula and in Malaysia

generally. The present mats and baskets of the Semang cannot therefore

be regarded as original Negrito productions. They have almost certainly

been adopted through contact with other cultures.

The same thing would seem to apply to the present basket-work of the

Philippine Negritos, of which however we know very little.

Netted bags of string are made by the women of both the Great Andaman

and the Little Andaman Divisions, and are used for carrying or storing

small objects such as shells, fruit, roots, etc. The string used for

these is made from the fibre of the Gnetum edule in the Little Andaman,

and is generally made from the same fibre in the Great Andaman, but in

the latter division the fibre of Hibiscus scandens is prized for this

purpose and is used instead of that of the Gnetum when it is available.

Small hand fishing nets are made by the women in both divisions from

thread of Gnetum fibre. In the Great Andaman Division netting is also

used for personal ornaments, the thread used for this purpose being

generally made from Anadendron fibre.

The mode of netting is always the same. Netting needles are used, made

from slips of bamboo or cane, and varying in size according to the work

for which they are intended. The netting string or thread is wound on

to this needle. For the foundation of the net a short length of string

is taken and the two ends tied together, the loop thus formed being

placed over the big toe as the woman sits on the ground.

The knot used in netting is that known as the fisherman’s knot, in use

all over the world. The needle is held in the right hand, is passed

from above downwards through the loop marked a in the figure, and is

drawn through far enough to leave a new loop b of the required size.

This loop is held between the thumb and forefinger of the left hand,

the uniformity in mesh being apparently obtained largely by the sense

of touch. The needle is then passed through the new loop b again from

above and is drawn out leaving a loop c, through which the needle is

passed once more, forming the finished knot, which is then drawn taut.

When netting is made for personal apparel it is generally ornamented

with strips of Dendrobium skin worked in with the thread in the course

of the netting. Netting for personal ornaments consists of a sort of

bag open at both ends. Each end is tied with a string passed through

the ultimate loops, and this string serves to tie the band of net round

the waist or neck or wrist or knee.

Fishing nets are attached to a handle. The handles of the Little

Andaman Division and those of the Great Andaman Division are different

in shape. The Great Andaman net with its hoop can be folded up

compactly for carriage, while that of the Little Andaman cannot.

The nets of rope formerly used in the Great Andaman for catching turtle

have been already described.

DOMESTIC IMPLEMENTS AND UTENSILS.

The Andamanese are perhaps the only people in the world who have no

method of their own of making fire. At the present time they obtain

matches from the Settlement of Port Blair, and a few of them have

learnt, either from Burmese or from Nicobarese, a method of making fire

by the friction of pieces of split bamboo. Formerly, however, they had

no knowledge of any method by which fire could be produced. Fires were

and still are carefully kept alive in the village, and are carefully

carried when travelling. Every hunting party carries its fire with it.

The natives are very skilful in selecting wood that will smoulder for a

long time without going out and without breaking into flame.

The most interesting of the Andamanese domestic utensils is undoubtedly

the cooking-pot. Pots are made in all parts of the Andamans where

suitable clay is to be found. The clay is obtained and is freed as far

as possible from stones and gritty matter with the hands. It is then

moistened with water and kneaded on a board consisting of a portion of

a broken canoe. It is worked very stiff and after kneading is rolled

beneath the palm of the hand into long thin rolls. These are built up

into a pot by coiling, the requisite degree of thickness being obtained

by pressure of the thumb and first finger. When the pot has been built

up to the required size and shape, the surface, inside and out, is

moistened and is scraped over in all directions with an Arca shell. The

pot is dried in the sun for a few hours and is then baked by placing

inside and around it pieces of burning wood. The pot often cracks in

the baking, and another has to be made.

In the North Andaman the pots are made with pointed bottoms, and are

generally small, the largest having a capacity of only six or seven

pints. They are made by women only. In the South Andaman the pots have

rounded bottoms and, on the average, are larger and thicker than those

of the North Andaman. They are made by both men and women, the best

being made by men. In the Little Andaman the pots are larger, and,

particularly, deeper than those of the South Andaman, and have somewhat

pointed bottoms. The pots of the Jarawa are similar to those of the

Little Andaman.

In the North Andaman pots are not ornamented, but in the South Andaman

they are decorated with usually simple patterns of dots and lines made

with a small pointed stick. The Little Andaman pots that I have seen

were not decorated.

The pots are used for cooking, i.e. for boiling meat and vegetables. In

the North Andaman small pots are specially made for melting the

composition used for covering the binding on the heads of arrows.

Neither the Semang nor the Philippine Negritos make any kind of

pottery. The origin of Andamanese pottery therefore is a problem of

some interest. It is almost certain that the early Andamanese were

acquainted with pottery before they were divided into the Great Andaman

and Little Andaman Divisions. One of the very few words which is the

same in all Andamanese languages is the word for pot, buču in the

Little Andaman, buǰ in Aka-Bea and peč in Aka-Jeru. The most reasonable

hypothesis would therefore seem to be that the Andamanese learnt the

method of making pottery by coiling before they reached the Andaman

Islands but after they had become separated from that part of the

Negrito race from which the Semang and the Philippine Negritos are

descended. Of course it is possible that the art of pottery may have

been an original possession of the Negritos and that the Semang and

Philippine Negritos may have lost it. A people could easily lose the

art if they were compelled in the course of their migrations to spend

three or four generations in a region that lacked clay suitable for the

purpose.

The Andamanese make buckets of wood, which they use for carrying and

holding a supply of water. A solid piece is cut from the trunk of a

soft-wooded tree and is hollowed out with a chisel made by attaching

the blade of an adze to a stick. The natural form of the wood is

retained, only the bark being removed from the outside while the inside

is chiselled out to leave sides of about a centimetre thick and a

bottom of somewhat greater thickness. In order to render the bottom of

the bucket water-tight the natives of the Little Andaman pour over it

on the inside melted bees’-wax. The natives of the Great Andaman use

for this purpose the same composition with which they cover the

bindings of their arrows, which certainly is superior to bees’-wax.

A bucket of the shape used in the Great Andaman is shown in Plate VII,

on the left. A strip of cane is attached round the middle of the

bucket, and to this in turn is attached another strip of cane which,

being passed across the front of the chest, enables the bucket to be

carried on the back.

The buckets of the Little Andaman are cut with thinner sides than those

of the Great Andaman. A strip of cane is fastened round the bucket near

the top. From this two other strips of cane are attached by both ends,

passing under the bottom, and a third strip is attached which passes

over the head and supports the bucket when it is carried on the back.

In the Little Andaman the outside of the bucket is charred with fire.

Occasionally pieces of the giant bamboo, which does not grow in the

Andamans, are found on the shore, having drifted from the Burma coast.

When a sound piece is found it is made into buckets each formed of a

single joint.

Water vessels are made from bamboos that do grow in the islands. A

length of bamboo of good diameter is cut, containing three joints. The

partition of the lowest node is preserved to serve as a bottom, and the

other partitions are broken through with an arrow. This is the usual

vessel used for carrying water on a journey by land or in a canoe, and

for keeping a supply of drinking water in the hut.

A single joint of the same kind of bamboo is used as a cooking vessel.

It is cleaned, washed, and dried, and is then filled with meat tightly

packed. The top is closed with leaves and the bamboo is placed on the

fire until the meat is cooked. Meat is not cooked in this way for

immediate consumption. It will keep for twenty-four hours or even

longer if not opened. To obtain the contents the bamboo is split open

with an adze.

Trays used for food are cut from soft wood of a species of Sterculia

with an adze. They are shallow and somewhat long and narrow, with

pointed ends.

A large Pinna shell is occasionally used as a tray or dish for holding

food, or for mixing clay with water for painting. A Nautilus shell

forms a fairly convenient cup or drinking vessel and is frequently used

for that purpose, as well as for baling out water from a canoe.

Bamboo tongs are made by bending double a piece split from a bamboo,

and cutting the ends to the required shape. They are used for lifting

from the fire anything too hot to be taken in the hand and are chiefly

of service in cooking.

Digging sticks are made from various kinds of wood, being simply

pointed at one or both ends. These sticks are not as a rule preserved,

but made as required with an adze and thrown away after use. They are

used for digging up edible roots.

Hooks for picking fruit, such as the Artocarpus, are made by attaching

a small piece of wood to the end of a bamboo. Hooks for catching crabs

are made from the wood of the Rhizophora by taking advantage of the

natural form where a small branch joins a larger one.

The fan-shaped leaf of a Licuala palm is made use of in many ways. The

edges of the leaflets are sewn together with fine strips of cane, and

the sewn leaf is then used either as a sunshade or umbrella for

protection from sunshine or rain, as a sleeping mat if the proper

article be lacking, as a screen to make the roof or sides of a hut more

wind or rain proof, as a wrapper for making objects of all sorts into

bundles, and as a winding sheet for a corpse.

Torches are made from resin, which is broken into small pieces and

wrapped up in leaves of a species of Crinum (? lorifolium), a few

pieces of smouldering charcoal being added before it is tied up. The

torch is then parcelled by marling with a strip of cane or a length of

some tough creeper. These resin torches are used in turtle-hunting and

fishing expeditions on dark nights.

Other torches are made of fragments of rotten Dipterocarpus wood. They

are used only in the village.

PERSONAL ORNAMENT; GREAT ANDAMAN.

At the present time the natives of the Great Andaman Division obtain

from the Settlement of Port Blair cloth which both men and women wear

round the loins. They also obtain beads from which they make necklaces.

The following is a list of the various personal ornaments made by the

natives of the Great Andaman Division, and worn by them.

Rope girdle. Every man wears some sort of girdle round his waist, and

this was formerly the only object that was constantly worn by men. The

girdle may consist of a length of rope of Hibiscus fibre or a length of

ornamental cord made by wrapping the yellow skin of the Dendrobium over

two strands of Hibiscus fibre and then twisting these into a two-ply

cord.

Necklaces and garters of string. Both men and women are often to be

seen with a simple piece of string, usually of Anadendron fibre, tied

round the neck or around the leg just below the knee.