July 15, 1905.] 
FOREST AND STREAM 
49 
tens of thousands of these reptiles every year. As his 
are, so far as I am aware, the only turtle farms in the 
world which are highly successful, a description of his 
establishment and methods will, I think, prove inter- 
esting and serve as a guide to those who may have 
similar undertakings in view. In passing I may re- 
mark that I have known Mr. Hattori these twenty 
years and have spent a number of summers on his 
original farm, collecting, with his kind consent, ample 
materials for my studies on the development of 
Chelonia. In return, Mr. Hattori is kind enough to 
say some of the facts and suggestions I have been able 
to give him, based on my embryological studies, have 
been of service in carrying out improvements. 
The Hattori family has lived a long time in Fuka- 
gawa, a suburb of Tokyo, which lies on the “Surrey” 
side of the Sumida River, and which, having been 
originally reclaimed from the sea, is low and full of 
lumber ponds, and until recently of paddy fields. The 
occupation of the family was that of collecting and 
selling river fishes such as the carp, the eel, and the 
crucian carp, and of raising gold-fishes, in addition to 
the ordinary farmer’s work. As far back as in the 
forties of the last century, the high price commanded 
by the “suppon” seems to have suggested to the father 
and the uncle of the present Hattori the desirability of 
cultivating it, and this idea, once started, seems never 
to have been lost sight of, although lying in abeyance 
for a long time. 
In 1866 the first large turtle was caught, and from 
then on additions were made by purchase from time to 
time, so that in 1868 there were fifteen, and by 1874 the 
number reached fifty, which were all very healthy, with 
a good admixture of males and females. In 1875 these 
were placed in a small pond of 36 tsubos,* with an island 
in the center which was intended for the turtles to lay 
eggs on. They, however, seemed to prefer for this 
purpose the space between the water edge and the outer 
inclosure; hence, to suit the tastes of the reptile, the 
pond was hastily modified into a form very much like 
the one in use at the present day. That year over 
one hundred young were hatched, but, unfortunately, 
they were allowed to enter the pond in which the adults 
lived, and all but twenty-three of them were devoured, 
making it evident that some means were necessary to 
In order to give a connected account oj the raising 
of tortoises, we might begin with a description of the 
pond for large breeding individuals, or “parents,” and 
with an account of egg-laying and hatching. 
The “parents’ ponds” does not differ in any remark- 
able way from the general plan of a pond given above. 
Usually one of the largest ponds is chosen, and it can 
be distinguished from the others, because one or two of 
its slopes are usually kept up very carefully, while the 
other slopes or those of other ponds are apt to be worn 
by rain and wind and to become rugged. These well- 
kept slopes are invariably on the warmer sides, where 
tlie sun pours down its midsummer rays longest, and are 
carefully worked over in the spring so that the tortoises 
will find it easy to dig holes in them. In the breeding 
season these sides are seen to be covered with wire 
baskets which mark the places where the eggs have 
been laid. 
Copulation takes place on the surface of the water 
in the spring. Egg deposition begins in the last part 
of May and continues up to the middle of August. Each 
female comes out of the water and wanders about a 
httle, while on the banks of the pond in search 
of a suitable locality in which to deposit _ eggs. 
ILaving finally chosen a spot, with her head directed 
up the bank she firmly implants her outstretched 
fore-feet on the earth, and during the whole opera- 
tion never moves these. The process of egg de- 
position, which takes altogether about twenty minutes, 
may be divided into three portions occupying about the 
same length of time, namely: (i) digging a hole, (2) 
dropping eggs in it, and (3) closing the hole. The 
digging of the hole is done entirely with the hind legs. 
Each with its nails outstretched is moved firmly from 
side to side— that is, the right foot from right to left 
and the left from left to right, and the two are worked 
in a regular alternation, while the body is swayed a 
little from side to side, accompanying the motion of 
the legs. The force put in the lateral pressure of the 
feet is so strong that the earth that has been dug out 
is sometimes thrown off to a distance of 10 feet or 
more, although the largest part of it is hepped up 
around the hole. Digging seems to be continued as 
long as there is any earth within the reach of the legs 
to be brought up. The result is a squarish hole with 
The traces of a spot where the snapping turtle has 
laid eggs are (i) the two marks made by the forepaws 
holding on to the earth during the whole operation, and 
(2) a disturbed place some distance back of the line of 
the forepaws where the hole has been made. The three 
marks are at the angles of a triangle. I have noticed 
a very interesting fact^in regard to these traces. When 
a young female is depositing her first eggs, she is very 
clumsy, the hole being badly made and the filling in of 
it very imperfect, so that often a part of it remains 
open. Old females are extremely neat in their doings, 
and one can determine at once the age and size of the 
female by the skill displayed and by the distance be- 
tween the three marks of egg deposition. This shows 
that although the elaborate actions necessary in egg 
laying must be, in the main, due to instinct, each in- 
dividual has to add its own experience to the inherited 
impulses and is able thus only to accomplish the desired 
end with perfection. 
In Hattori’s farm a person goes around the “pa- 
rents’ pond” once a day or so and covers up with 
wire baskets all the new deposits made since the last 
visit. Each cover basket may be marked with the date if 
necessary. This covering serves a two-fold purpose — 
the obvious one of marking the place, and in addition 
that of keeping other females from digging in the same 
spot. When hundreds, or even thousands, of these 
baskets are seen along the bank of a “parents’ pond,” 
it is a sight to gladden the heart of an embryologist, to 
say nothing of that of the proprietor. 
The hatching of the eggs takes, on an average, sixty 
days. The time may be considerably shortened or length- 
ened according to whether the summer is hot and the sun 
pours down its strong rays day after day, or whether 
there is much rain and the heat not great. It may become 
less than forty days or more than eighty days. By the 
time the last deposits of eggs are made in the middle of 
August, the early ones, which were laid in May or 
June, are ready to hatch; and inasmuch as if small tor- 
toises that have just emerged from the eggs are allowed 
to get into the “parents’ pond” they are devoured by 
their unnatural fathers and mothers, a special arrange- 
ment has now to be put up to prevent this. The left 
side of the plan in cut 2 are intended to show this 
arrangement. Long planks about 8 inches wide are put 
RIVER 
Cut 1,— Plan of a turtle farm.' 
Cut 2.— Plan of pond. 
protect them from their untiktural parents. ^Thus was 
gradually evolved the present system of cultivation. 
In general appearance a turtle farm is at a first glance 
nothing but a number of_ rectangular ponds, large and 
small, the large ones having a size of several thousand 
tsubos. The ponds are undergoing constant modifica- 
tion, being united or separated just as need arises, so 
that their number may vary considerably at different 
times. Cut t gives the plan of the Hattori turtle farm 
at Fukagawa as at present laid out. There pass through 
the farm two small canals which communicate on the 
one hand with the river across the road, and on the 
Other with the ponds, so that the water can be drawn 
into, or emptied from, each of them at will. 
All the ponds, whether large or small, are constructed 
very much on the same plan. They are limited on their 
four sides by plank walls, the top of which may either 
be on the level of the ground (see the right side of the 
section, cut 2) or may be more than a foot above the 
ground when two ponds are contiguous (the left side, 
cut 2). In either case the plank wall has a cross plank 
of some width at right angles to it on its top, and is 
also buried some inches in the ground. The former ar- 
rangement is, of course, to prevent the tortoises from 
<climbing over the wall, and the latter to prevent theirr 
from digging holes in the ground and making their 
■escape in that way,, while at the same time it serves to 
■exclude the moles. On the inner side of the plank wall 
Hiere is more or less of a level space, and then a do^yn- 
ward incline of 3 or 4 feet. At the foot of this incline 
and directly around the water’s edge there is another 
level space which enables people to walk around the 
pond. From the edge of the water the bottom of the 
pond deepens rather rapidly for a space of some three 
feet, and there reaches the general level of the bottom, 
which is about two feet below the level of the water. The 
greatest depth of a pond is about three feet and is always 
toward the water gate by which the pond communicates 
with the canals. The bottom is of soft, dark mud, several 
inches thick, into which the tortoises are able to retire to 
pass the winter. 
On a turtle farm one or more of the ponds is always 
reserved for large breeding individuals, or “parents,” 
;as they are called. The just-hatched young or the first- 
year ones must have ponds of their own, as must also 
fhe second^^year ones; those of the third, fourth, and fifth 
years may be more or less mixed. 
'■One tsubo, an area 6 feet square, is the tmit in the rneasure- 
,-%!ent of small land surfaces. 
the angles rounded off, and although its size differs 
with the size of the female, it is generally about 3 to 4 
Inches across at the entrance, with the depth and width 
inside about 4 inches or more. When digging is 
finished eggs are dropped from the cloaca into the 
hole, which naturally lies just below it. The eggs are 
heaped up without any order, but, there being no 
chalazae, the yolk is able to rotate in any direction, and 
the blastoderm, having the least specific gravity, always 
occupies the highest spot of the yolk in whatever 
position the egg may happen to be dropped. The eggs 
are generally spherical in shape, although sometimes 
more or less oblate. Their diameter is in the neighbor- 
hood of 20 millimeters, the largest being as large as 
24 millimeters, the others smaller according to the size 
of the females. The number of eggs in one deposit 
varies from 17 or 18 to 28 or more, the smaller indi- 
viduals producing the smaller number. 
When the eggs have all been deposited, the turtle’s 
legs are again put in requisition, this time to fill up the 
hole, which is done by alternate motions as before. 
The earth about the hole is used at first, but search is 
made for mere loose earth for a little distance, as far 
around as the legs can reach with a slight motion of 
the body either to the right or left without moving the 
front legs. Toward the end of the process the loose 
earth is trampled down. When the hole is well filled 
up to the level of’the ground, the turtle turns around 
and goes immediately down into the water, not casting 
even one backward glance. 
I have noticed an interesting contrast between the be- 
havior of Trionyx and of Clemmys during the egg de- 
position. If one wants to watch a Trionyx depositing 
eggs, one has to crawl on all fours behind the plank 
wall of the pond and peep through a hole, being care- 
ful not to show himself. The moment the snapping 
turtle sees anyone, it stops in whatever part of the egg- 
laying process it may be engaged and plunges straight 
into, the water. Utterly different is the behavior of 
Clemmys. When once it begins the process of egg-lay- 
ing it is never deterred from carrying it out, no matter 
how near or how boldly one may approach. Whenever 
I watched a Clemmys working away in the direct mid- 
summer rays \yith its carapace all dried up and with its 
eyes alone moist. I could not help comparing it to 
a slave of duty fulfilling his fate with tears in his eyes. 
What causes such a difference of behavior in the two 
species? What is its significance? What difference in 
the nervous system corresponds to it? 
up lengthwise around the edge of the pond, leaving 
perhaps i foot margin between them and the water. 
The two successive planks are not placed contiguous, 
but a space of about 3 feet is left between every two, 
and closed by a bamboo screen put up in the shape of 
an arc of a circle, with its convexity toward the pond. 
Thus the slope or the bank where the eggs have been 
deposited is completely cut off from the pond itself. 
In the center of every pocket-like arched space made 
by a bamboo screen an earthenware jar is placed with 
its top on the level of the ground, and some water is 
put in it. This elaborate arrangement is for the recep- 
tion of the young tortoises, which, as soon as they 
break through the egg shells— those belonging to the 
same deposit generally coming out at the same time — 
crawl up to the surface of the ground by a hole or h<des 
made by themselves, and go straight down the incline 
toward the pond,, as naturally as the duckling takes to 
the water. They are stopped, however, in their down- 
ward hydrotaxic course by the planks put up, as stated 
before, around the pond, and they crawl along the 
length of the planks and sooner or later drop into the 
jars placed in the recesses between every two planks. 
A man going around once or twice a day can easily 
collect from these jars all the young hatched since the 
last visit. 
The young just hatched are put in a pond or ponds . 
by themselves and given finely chopped meat of a fish 
like the pilchard. This is continued through September. 
In October Trionyx ceases to take food, and finally 
burrows into the muddy bottom of the pond to hiber- 
nate, coming out only in April or May. The young 
are called the first-year ones until they come out of 
their winter sleep, when they are called the second- 
year young. At first the same kind of food is given 
these as that given to the first-year young, but gradu- 
ally this may be replaced by that given to older indi- 
viduals, namely, any fish meat or crushed bivalves, etc. 
Figure 2, plate I., shows a lot of the second-year young 
in August. From the third to the fifth year, inclusive, 
the young need not be kept in ponds_ strictly according 
to age, but may be more or less mixed, if necessary. 
The young of these years are also the best and_ most 
delicate for eating and are the ones most sold in the 
market. In the sixth year they reach maturity and may 
begin to deposit eggs, although not fully vigorous till 
two or three years later. How old these snapping 
turtles live to be is not known. Those i foot and more 
jn length of carapace must be many years oW. 
