300 General Notes. = [March, 
also over twice the diameter of those of Isotoma, which measure 
only about .15 of a millimeter in diameter, so that there is more 
yolk present and development is much more extremely mero- 
blastic or decidedly epicyemate in character. This will be evi- 
dent if Figs. 9, 10, 11 and 12 are compared with the earliest 
stages of Isotoma figured by Packard in the memoir just cited. 
One feature in the development of Anurida which has inter- 
ested me greatly is the presence of a very rudimentary spring or 
elater, e/, shown from below in Fig. 4 and from the side in Fig. 3. 
This appendage, which probably represents a pair of degenerate 
limbs, is produced from the anterior, inferior part of the fourth 
abdominal segment, but on the ventral side of the adult no sign 
of its presence is visible, as may be gathered from an examination 
of Fig. 1. This organ in Anurida does not arise from the penul- 
timate segment, as in Isotoma, as stated by Packard, but from the 
antepenultimate or fourth abdominal segment which is in reality 
the one from which the elater arises in such genera of Collem- 
bola, as Lepidocyrtus, Triana, Tomocerus, etc. In the just- 
hatched larval Anurida, the elater is developed to exactly the 
same degree as in Triena mirabilis Tullb., according to Brook.’ 
The inference, therefore, is that the springless genera of Collem- 
bola are degenerated forms which have descended from others 
which were provided with well-developed elaters. In fact it 1$ 
now possible to trace the gradual degeneration of the elater 
through the genera Achorutes, Xenylla, Trizna and the young 
of the species under consideration here. Linking this series with 
those having a more developed elater and tenaculum, and these 
again with such forms as Campodea and Machilis, we réalize what 
a remarkable series of differential changes the abdominal appen- 
dages of the Thysanura and Collembola have undergone, starting 
probably from the still less modified Symphyla, in which there is 
no differentiation even between the appendages of the thorax and 
abdomen. 
The earliest stages were not considered, as the ova were too 
on _ Opaque to be studied by transmitted light, and the earliest intima- 
tion of the formation of the germ is shown in Fig. 8 at gå, the germ- 
eing viewed in profile transversely or from one end. The 
1-band or ventral plate forms a pronounced thickening which 
= lies on one side of the vitellus, with its longest diameter coim- 
 ciding with the longer diameter of the egg. This germinal band 
soon becomes widest anteriorly, as shown by the transverse pro- 
file views of it represented by Figs. 11 and 12. From these ar 
~ lateral profile view (Fig. 10) of the same stage, I have con- 
__ Structed the diagrammatic representation (Fig. 9) of the germ- 
band as it would be seen by a 
object, extending over very nearly a semicircumference of the - 
reflected light, or as an opaque 
re 1 Journ, Linn, Soc. London, XVII, 1882, pp. 21-22, pl. 7, figs, 11 and 12. 
