10 PROCEEDINGS OF THE NATIONAl, MUSEUM. voi.. 60 



maxillipeds are chelate. In the other genera, when one of these ap- 

 pendages is chelate, the other will be found to be armed with simple 

 claws or often only with setae. Lamproglena^ as noted above, is the 

 chief exception, and here the second maxillae and maxillipeds have 

 apparently changed places. The preceding table has been compiled 

 for the purpose of contrasting the structure of the antennae and the 

 mouth parts, with the exception of the mandibles, in the various 

 genera. The available data on the structure of the mandibles at 

 the present time is too meager to warrant their inclusion in the table. 

 Swimming legs. — The swimming legs vary greatly in number and 

 structure. Most of the genera have four pairs, but in Eudactylina, 

 Nemesis^ Lainproglena^ and Donusa there is a fifth pair. In the 

 first three the fifth leg consists of a single laminate process, but in 

 Donusa each fifth leg is biramose and the rami are three jointed, like 

 the other four pairs. In addition to the genera just named the first 

 four pairs of legs are also biramose in C ongericola., Pagodina. Kr0ye- 

 ria, Peniculisa, and Sagtim, nine genera in all out of the twenty in 

 the family. In Donusa and Kr0yeria each ramus is three jointed, 

 but in the other genera the number of joints varies, no two genera 

 being alike. In Bassettithia the first three pairs of legs consist of 

 basal joints only, while the fourth pair are biramose, with one- 

 jointed rami. In Cyhieola and Ventriculina there are only three 

 pairs of legs, the first pair biramose, with one- jointed rami, the 

 second and third pairs uniramose. In Pseudoclavella the first two 

 pairs are biramose, the third and fourth pairs uniramose. In 

 Pseudocycnus the first and fourth pairs are uniramose and are made 

 up of the basipod only, the second pair is biramose, the rami one 

 jointed, and the third pair is uniramose and two jointed. In 

 Hatschehia and Caetrodes only the first two pairs of legs are present. 

 each biramose and the rami two jointed. In Dichelesthium the first 

 two pairs are biramose and the third pair uniramose, while the fourth 

 pair is entirely lacking. In Lernanthropus the first two pairs are 

 biramose with one-jointed rami, while the third and fourth pairs 

 are modified into long cylindrical processes, apparently concerned in 

 respiration. In Anthosoma the first three pairs are replaced by 

 foliaceous plates and the fourth pair is lacking, while in Norion 

 apparently only the first pair is present, and these are uniramose and 

 one jointed. In the following table these differences are brought out 

 clearly. The most noticeable feature is the prevailing lack of uni- 

 formity. The absence of fifth legs is the only character that ap- 

 proaches regularity and even there twenty percent of the genera 

 show exceptions. It almost seems as if the various characters must 

 have been shuffled for each genus separately in order to produce such 

 prevalent and radical discrepancies. 



