AM0RPH1N7E. 387 



rising in the middle of a depression. In S. ocellata the depression is hardly to be 

 detected. In M. tiliae these markings cannot be always observed owing to gum, 

 but in one specimen, in consequeoce or in spite of the gum, the ridges dividing the 

 cells were broad and flat, and left a depression marking each cell with a small, central, 

 raised boss. The micropyle was not seen in the egg of M. tiliae. In Amnrpha and 

 Smerinthus it is a small rosette of about 20 narrow cells some -04mm. in diameter, 

 apparently with a few smaller central cells. It is to be noted that the narrower end 

 of the egg is the nadir, the thicker the micropylar end. The measurements are — 



A. populi. S. ocellata. M. tiliae. 



Greatest length .. .. .. 2*o5mm. 2*05111111. i-o^mm. 



Greatest breadth ;. .. r"omra. i*55mm. r55mm. 



Lesser breadth (towards nadir) . . i*65mm. • 



Greatest height . . .. .. i*6omm. i*30mm. i*35mm. 



Lesser height (towards nadir) . . 1-451111x1. 1*151x1111. riomm. 



Diameter of cells of sculpture .. '022mm. *02mm. -02mm. 



There is- some probability that the peculiar character of the sculpturing of the egg 

 surface is due to the cement covering hardening and contracting in peculiar relation 

 to what is probably the true sculpturing, viz., a simple fine network, the fine lines 

 of the network attracting some of the cement, and the remainder aggregating in the 

 centre of each cell. These eggs present what is an almost invariable character of flat 

 eggs, viz., a depression of the upper surface, slowly increasing from day to day as 

 evaporation goes on. That of S. ocellata seems to yield most readily to this change, 

 that of M. tiliae comparatively little, a difference no doubt due to the great thick- 

 ness of cement that clothes the latter, acting as a great check to evaporation 

 (Chapman). 



Larva. — Newly hatched : The larvae of these three species are extraordinarily 

 alike and the likeness goes into much detail. Not much short of ^in. long, when they 

 stretch themselves, and with large green heads, they are almost identical in size and 

 outline as well as in colour. A. populi is of a paler, whiter green. To the naked 

 eye they are distinguishable by the colour of the horn. In A. populi this is a 

 green almost the same as the rest of the larva. In S. ocellata a pinkish-red, and 

 in M. tiliae a dark fuscous shade, seen with a glass to be more marked on the upper 

 and under aspects of the horn. A sharp eye may also detect that the larvae of 

 Amorpha and Smerinthus have already the lateral oblique lines distinguishable 

 as of a slightly different shade of green, whilst that of M. tiliae has no trace of them, 

 but has a broad subdorsal and a not quite so broad lateral longitudinal stripe. All 

 have 8 subsegments to the abdominal segments, and all have forked hairs. In 

 these there is some little difference in each species. There is also a difference 

 in the crochets of the prolegs. Three larger hairs of the same forked character 

 as the minute ones mark tubercles i, ii and iii, the latter being the least obvious ; 

 they are on subsegments 3, 6 and 4 respectively. On the 2nd and 3rd thoracic 

 segments there is a subsegment rather broader than the others, and terminating by a 

 sulcus just above the spiracular region. This is the 3rd or 4th subsegment of the 

 5 or 6 which seems to be the number of subsegments on these segments. This 

 subsegment carries three tubercles which are in line with i, ii and iii. The first 

 two (i and ii) carry two forked hairs, one in front of the other, the third (iii) carries 

 only one. These are substantially the same in all the larvae, but are, perhaps, a 

 little more easily seen in M. tiliae. The crochets seem to be 9 in number usually, 

 they are of a much darker brown in A. populi than in the others. The forked 

 hairs differ in the three species. In all, the larger hairs have the forks less pro- 

 nounced, the smaller hairs vary much in size in the same individual in different 

 places, so that it is difficult to say that they differ in size in the different species, 

 but the forked extremity has a specific form in each case. In S. ocellata there 

 is very little enlargement at the point of forking, the hair being" little more than 

 bifurcate, the two branches leaning at an angle of 120 , and being between \ and 

 \ the length of the hair in length, and tapering regularly to a tine point. In A. 

 populi the branches are rather shorter, spread more laterally, and have the angle 

 between them so filled up as in some degree to make them resemble a fish's 

 tail, especially as the central thin lamella is marked with radial lines, and even 

 slightly notched at the margin. In M. tiliae the two branches spread nearly at 

 right angles to the stem, and so there is little room between them to fill up, 

 yet there is a central portion that actually projects at the angle between the 

 branches, and is marked by a few radial lines. The branches are perhaps proportionally 



