ON THE AGAMIC REPRODUCTION AND MORPHOLOGY OF APHIS 63 
downwards, and then sharply upwards and backwards, to a little 
beyond the line of the posterior edge of the procephalic lobes. 
The whole of this portion of the blastoderm belongs to the head. 
In the re-entering angle between it and the labrum the mouth is 
placed ; it is a small aperture, whence the cesophagus can be traced 
ascending and passing backwards with a gradual curve. 
Behind the cephalic region, the thoracic blastoderm passes nearly 
horizontally backwards, and already presents traces of a division into 
its three somites. Its upper surface is close to the pseudovitelline 
membrane, and consequently is covered by but a very thin layer of 
yelk-like granules. 
At the end of the rudimentary thorax the blastoderm is suddenly 
folded forwards, so that the sternal surface of the hinder part of the 
future abdomen is almost in contact with that of the thorax. Having 
come opposite the anterior edge of the thorax, it is bent backwards, 
at right angles to its previous direction, for a short distance,—the 
extreme end being finally folded parallel with this part, and with its 
apex towards the head. 
The great mass of the yelk lies over the abdominal blastoderm, in 
the space left between it and the pseudovitelline membrane. The 
appendages present a singular and beautiful uniformity. No trace 
of the pigment of the eyes is to be seen. The next anterior, pair of 
appendages (@ 7) are more slender and elongated than the others, and 
are bent inwards near their base so as to forma sort of elbow. In 
consequence of this, their terminal portions are more approximated 
than their distal ones, and lie close together and parallel, These 
appendages are the antennz; and it is worthy of remark, that they 
arise from the procephalic lobes, or from the point of junction between 
them and the rest of the cephalic blastoderm above the mouth. 
Behind these and behind the mouth (though the anterior pair are 
very close to that aperture, and might even be described as more or 
less lateral in relation to it) are three pair of short, similar, conical 
processes. Of these the anterior pair (IV’) are the largest, and are the 
mandibles; the two other pairs are nearly equal: the anterior (Vv’) 
represent what are ordinarily termed the maxilla, but which might 
be more properly called “first maxilla,” since the second pair (VY), 
which eventually give rise to the so-called “labium,” are precisely 
like them, and, as Zaddach (fc. zzfra) has shown, fully deserve the 
title of “second maxilla.” 
Three pairs (VII, VII’, IX’) of short processes, unjointed and not 
much longer than the trophi, represent the thoracic limbs. 
The abdomen presents obscure traces of a division into segments. 
