46 ON THE AGAMIC REPRODUCTION AND MORPHOLOGY OF APHIS 
keime betrachten), we may assume that these dAp/zdes were capable 
of bringing forth more than eight ova.” 
Von Siebold then goes on to describe the colleterial glands, and 
the spermatheca, which had not before been seen. If the ovaries of 
Aphis Loniccre are not constructed on a totally different plan from 
those of the species I have described, it is, I think, pretty clear that 
Von Siebold, like Morren, has mistaken the ovarian glands for the 
rudiments of the ova. Indeed, his phraseology indicates that he 
himself had no great confidence in his interpretation of the parts. 
$4. The Development of the Pseudovartuin 
In the viviparous female, the germ increases in size, and gradually 
becomes separated from the terminal chamber by the successive 
development and separation by constriction of new pseudova. The 
number of chambers between the terminal one and that nearest 
the vagina, therefore, varies until it attains its maximum, which is 
necessarily regulated by the ratio between the time required for the 
perfection and birth of a larva and the rate at which new pseudova 
are detached from the pseudovarium. In the species of Aphzs which 
I examined, I found ordinarily four or five such chambers. Germs 
between zppth and 4} 5th of an inch in length presented the following 
characters (Pl. NNAVII. [Plate 2] fig. 1):—They exhibit a central 
darkish matter, surrounded by a clear cortex. The latter is composed 
of a single layer of a substance similar in appearance to that com- 
posing the mass of the germ above described, while the central 
substance is obscured by a number of minute granules which hide 
its internal structure. Nevertheless, I have occasionally detected 
what I believe to be endoplasts, scattered through its substance, as 
in Pl NNNVIIL. [Plate 2] fig. 1, which represents a germ in this 
stage treated with very dilute acetic acid ; and as in a more advanced 
condition we shall find such bodies easily recognizable, I do not 
doubt that the central substance has the same fundamental compo- 
sition as the peripheral layer. The central mass, it will be observed, 
complctely simulates the vitellus of an impregnated ovum ; and I will 
therefore term it a * pseudovitellus.” The peripheral clear layer is, on 
the other hand, in all essential respects comparable to a blastodermic 
vesicle; and I see no reason why it should not be called a blasto- 
derm, since the term is not necessarily confined to the product of 
impregnation. 
In a more advanced condition (fig. 3), the blastoderm has become 
thicker in all parts, so as to consist of at least two-or three layers of 
