March 1, 1901. | 
Supplement to the " Tropical Agriculturist. 
652 
Some larave^are legless or with the mere rudi- 
ments of a pair cf legs on each of the three rings 
behind the head, fleshy, smiillest at the tuil, and 
furnished with distinct hend and jaws ; such are 
some kinds of Beetle — and Wasp-grubs ; others are 
etroug and. fat, a few inches in length, with three 
pairs of legs we!! developed — as the Cockchafer 
grub. 
The caterpillars of the Butterflies and Motlisare 
often beautifully marked, and liave for the most 
part a pair ot articulated feet on each of the three 
segments behind the head, and pairs of flesliy 
appeudiiges called sucker-feet on some of the or,lier 
segmentii and at the end of the tail, not exceeding 
sixteen in all. These "Sucker-feet" enable the 
caterpillars to liold flrmly to the twigs they fre- 
quent. Proceeding onwards still by number of feet, 
the caterpillars of the Sawflies will be found in 
every case (Corn Sawfly, C. pygnioeus, excepted) 
to have, besides the thrte pairs of true feet, five, 
six, or seven pairs of sucker-feet, and also the 
pair at the end of the tuil (known as the caudal 
pro-leg). 
Sometimes, as with Grasshoppers, Locust, Aphi- 
des or Green Fly, Plant Bugs, &c., the young in 
the first stage — whether produced alive or hatched 
from the egg — much resembles the parent, that is, 
has a distinct insect siiape, of head witli horns, 
trunk or thoiax, furnished with six legs, and 
abdomen ; and differs mainly in size and in being 
wingless; but, whether in tliis shape, or what is 
knov?n asgrub, maggot, or caterpillar, or whatever 
kind of insect it may belong to in this stage, it is 
scientifically A. Lama. 
In this larval stage the insect feeds voraciously 
and often grows fast: the skin does not expand 
beyond certain limits, and when this pointis arrived 
at, the larva ceases feeding for a while ; the skin 
loosens, cracks, and is cast of by the creature 
inside, which comes out in a fresh coat, sometimes 
like the previous one, sometimes of a different 
colour or differently marked. This operation is 
known as moulting, and occurs from time to time 
till the larva has reached its full growth. The 
duration of life in the first or larval state is various ; 
iu some instances it only extends over a week or 
two; in some (as with the wireworm and cater- 
pillar of the goat moth) it lasts for a period of 
three, four, or five years. 
As far as observations go at present — that is to 
gay, with such kinds as have at present been 
obscured — larvae are not injured by an amount of 
cold much beyond what they are commonly called 
on to bear in this country, but they are liable to 
injury from over supply of moisture, whether from 
sudden ruin in warm weather or from full flow of 
sap of their food-plaut, and in tliis point of their 
constitutions we have a principle that may help 
much towards getting rid of them. 
When the larva has reached its full growth it 
ceases feeding, and (in the forms known as cater- 
pillar, grub, or maggoi) it usually either goes 
down into the ground and forms a cell iu the 
earth, or spins a " cocoon " (that is, a web) round 
itself of threads drawn from the lower lip (as in 
the well-known silkworm-cocoya), or in some way 
it makes or seelcs a shelter in which it changes 
from the state or larva to tliat of pupa. Tliese 
various changes are not mere matters of curious 
enquiry, buc cati be used very serviceably, in 
prevention of recurrence of attack. 
COMPOSITION OF INDIAN '<JO\VS' AND 
BUFFALOES' MILK. 
Agricultural Ledger (Calcuttj), No. 19, treats 
of this subject. 
The information on the composition of Indian 
Cows' and Buffaloes' Milk is very raeigre. One 
analy.sis of cows' and one of buffaloes' milk was 
published in the Proceedings of the Madras Board 
of Kevenue in 189l> (i:ide Board's Proceedings 
No. 387, dated 25tii May, 1892), and the average 
composition of each kind of milk is published in 
Mr. Collins' Agricultural Chemistry, page 28. 
Tiie latter were deduced from a number of analyses 
which Mr. Collins made at Poona, but unfortun- 
ately the Note Books containing the details wer« 
lost in transit by rail in 1898. 
The composition of cows' milk has been found 
iu England to be very regular for different breeds, 
and to possess a relationship between the several 
component parts for all the breeds. It is of 
importance to know whether these relationships 
hold good for Indian breeds of cows, and to what 
extent they vary iri the case of the buff.ilo. 
The analyses publi-hed by the Madras Board of 
Revenue are only two in number, and are there- 
fore quite insufficient for the purpose in any case. 
In addition, however, they show the proportions 
of Proteids and Lactose to be alltogether different 
from anything met with in the case of English 
cow's milk. 
Mr. Collins found, on the other haiid, that the 
composition of the milk of the several breeds of 
cows at Poona coincided with that of English 
breeds. The milk of the buffalo he discovered 
-was characterised by a high proportion of butter- 
fat and considerably more Proteids than that of 
the cow. Owing to the loss of the detailed note?, 
it seemed to me very desirable that the work 
should be repeated, and in the following pages I 
give the composition of the milk of some of the 
cows and buffaloes at Poona, and that of the cows 
at the SaidapeO Agricultural College Farm. This 
information is set out in the accompanying State- 
ments Nos. I., It. and III. 
It has been found in England that (a) there 
exists a relationship between the Solids-not-Fat, 
the fat and the specific gravity, and (6) between 
the proportions of Proteids, Lactose and Mineral 
matter (ash), and that these relationships are 
constant, within certain limits, for all the several 
breeds. 
Eegarding the former, several formulae exist. 
That of Eichmond {vide "Analyst," Vol, XX., 
p. 57) is expressed, thus: — S ='25G+ '2F+ -14. 
Another recently worked out by Leonard is 
expressed :-F = % (T— o) -f- (O'S— 0-004T— 
0-OlG) where T = Total Solids ; S = Solids-not- 
fat; F= Fat; G= the excess of gravity over 
1,000. The latter forinula gives results which differ 
only slightly from Richmond's. The proportions of 
