22 ILLINOIS BIOLOGICAL MONOGRAPHS [142 



end opposite the blastopore the ectoderm thickens and forms a second 

 invagination, that of the proboscis. The entire proboscis develops from 

 ectoderm except for a few mesenchyme cells which have migrated into it 

 to form the muscles. The blastopore becomes nearly closed and the an- 

 terior end of the intestine does not communicate with the cavity of the 

 proboscis during the embryological stages. 



The description of the larva also can be given only in the most general 

 terms at this time. The larva of Gordius robustus dilffers greatly in form 

 from that of Paragordius varius, but in essential structures the two do not 

 seem to differ much. The newly hatched larva (Fig. 20) is very much 

 elongated, but after a week or so it has become shrunken to about half its 

 original length (Fig. 14) and has increased slightly in diameter. Like all 

 other known Gordius larvae it consists of proboscis and body. The pro- 

 boscis is armed in front with three retractile stylets and at the sides with 

 three circles of hooks which point backward when the proboscis is extended, 

 but are withdrawn into the proboscis when this is retracted. A set of retrac- 

 tor muscles is inserted at the base of the stylets and protractors lie close 

 to the outer wall of the proboscis. The body in the newly hatched larva 

 is at least twice as long as the proboscis and of a slightly smaller diameter. 

 Both body and proboscis are covered by external, more or less irregular 

 cuticular rings which do not seem to be in any way related to the deeper 

 structures of the larva. The posterior end of the body runs out to a point 

 resembling a heavy spine. Between the proboscis and the body there 

 appears to lie a partition separating the end of the intestine on one side from 

 a cord of cells coming from the base of the stylets on the other. Just behind 

 this partition is a mass of cells belonging to the intestine which farther 

 back has very thin walls and encloses within its lumen two elongated masses 

 of a homogeneous, highly refractive substance. Montgomery figures 

 similar masses in the intestine of Paragordius varius and regards them as 

 excretory waste. In Gordius robustus these masses are later absorbed as 

 large cells invade this region. The intestine opens at the posterior end on 

 the ventral side somewhat anterior to the spinelike elongation. 



Beginning near the proboscis and extending backward two-thirds of the 

 length of the body on the ventral side are two rows of nuclei indicating the 

 rudiments of the ventral nerve cord. Only longitudinal muscles appear to 

 be present in the body and they adhere so closely to the outer wall that it 

 is difl5cult to detect them. 



When fully developed the larva ruptures the egg-membrane and escapes 

 from the egg-string by means of the armature of its proboscis. Quite active 

 at first, it becomes more and more sluggish as it grows older. Larvae 

 picked up with a pipette from the bottom of the dish containing the eggs 

 usually show only a few active forms. If among a number of larvae in an 

 open drop on a slide about ten per cent are active, then when a coverglass 



