NO. 1788. NORTH AMERICAN ERGASILID^— WILSON. 307 



mandibles can be reduced so much as to be unserviceable for chew- 

 ing. He gave preference to Thorell's classification, yet admitted that 

 the facts upon which it was founded are not reliable. But, like most 

 of the critics, he contented himself with picking the old classification 

 in pieces and made no effort to modify or improve it. 



In the following year (1871) Sumpf described another new genus, 

 Tseniacanthus, belonging to this family of Ergasilidge, and also gave a 

 second thorough revision of the mouth-parts, comparing them care- 

 fully with those in the genera Corycaeus, SappMrina, and Lichomolgus. 

 He agreed with Claus that both mandibles and maxillse are present, 

 and hence declared himself against Thorell's classification. 



He showed that Claparede, by some unaccountable error, mistook 

 the upper lip for a pair of fused mandibles, while the structure desig- 

 nated as the maxillary plate was manifestly the true mandibles. He 

 then endeavored to show that the Chondracanthidae, which Thorell 

 had included in his group of Poecilostoma, really belong with the 

 Siphonostoma by reason of their limited development, the fixed 

 parasitism of both sexes as well as their sexual dimorphism, and by 

 their degenerate body form and mode of life. 



Sumpf thus completed the tearing in pieces of Thorell's classifica- 

 tion which Claus had begun and Claparede had continued. But, like 

 those authors, he offered nothing in its place, so that, as far as the 

 Ergasilidse are concerned, we are left without any arrangement at all. 



In 1877 Carl Vogt published his " Recherches Cotieres;" the second 

 section of the second memoir is devoted to the family Chondra- 

 canthidse. In it he described a species of Ergasilus, which he named 

 E. rmigilis, but he believed it to be identical with Hesse's Mega- 

 hrachinus suboculatus, in which case the latter specific name would 

 take precedence (see p. 346). He then called attention to the close 

 resemblance between the free-swimming family Corycaeidse and 

 the Ergasilidse. After a very clear cut and logical comparison of 

 these two families with the Chondracanthidse, he closes with these 

 words : 



All these facts authorize us to conclude that the Corycseidse are the free forms corre- 

 sponding to the parasitic Ergasilidse, and are less degenerate, while the retrograde 

 metamorphosis has reached its maximum in the Chondracanthidae, especially in the 

 females. In this manner these three families, well distinguished by secondary char- 

 acters, constitute in reality but a single series, which reflects in its successive trans- 

 formations the analogous phases which the ancestral copepods must have passed 

 through in their passage from a free life to one of parasitism (pp. 99 and 100). 



Thus Vogt does not agree with Sumpf that the Chondracanthidse 

 really belong with the Siphonostoma. He takes the more logical 

 view that they bear the same relation to the Ergasilidse and Cory- 

 cseidse that is borne by the Lernseidse to the Dichelestiidse and Cali- 

 gidse. This seems to be by far the most sensible statement made with 



