442 Prof. K. Grobbcn on the Genealogy and 



number (20)* of the segments of tlie body as compared with 

 Avhat we find in Apus and Esfhen'a, the sliarp separation 

 between tlie thoracic and the abdominal region, tlie latter of 

 which is devoid of appendages in the adult, and the modifica- 

 tion of the second antenna into a lanielliform structure in the 

 female and into an apparatus of considerable size in the male. 

 By the absence of the shell, the elongated form of the body, 

 the powerful development of the musculature of the trunk, 

 which comes into action in the darting motion of the body, 

 and the not very large number of segments in the thorax and 

 abdomen, Brancliipns is characterized among the Euphyllo- 

 poda as the form best adopted to the movement of swimming. 



On com]iaring Apus with Branchipus the first point which 

 will have to be noticed as a primitive character in the case of 

 the former is the larger number (83) f of the segments of the 

 body. In Apus, too, the transition from the thorax to the 

 limbless abdomen appears to be a more gradual one, owing to 

 the fact that there is a continuous and very striking diminu- 

 tion in the size of the tlioracic ap])endages towards the rear. 

 As a peculiarity which is found in Apus alone among the 

 Euj^iyllopoda must be mentioned the large number (G'i) of 

 the thoracic appendages ; the eleven anterior segments of the 

 body each bear one pair of appendages (they correspond to 

 the eleven limb-bearing thoracic segments of Branc/iipus)^ 

 while upon the following seventeen segments we find a larger 

 and posteriorly increasing number of limbs, so that to the 

 last two limb-bearing segments there together belong twelve 

 pairs of appendages. I will not here attempt to decide whether 

 in the posterior section of the thorax of Apus we have to 

 deal with a fusion of several segments to form larger annuli, 

 or with a multiijlication of the appendages within the 

 segments, although I rather incline to the latter view. In 

 any case, whether concentration of segments or multiplication 

 of the a])))endagcs has taken place, it would be a question of 

 a secondary condition. 



Thus, should the actual number of the body-segments be 

 determined by the number of the appendages, the segmenta- 

 tion of the body, in this case unusually extensive, would 

 surely have to be regarded as a secondary character. 



Ajnis bears a small shield-sha])ed shell covering the ante- 

 rior segments of the thorax, and herein possesses, as opjiosed 

 to JJranchipus, an old character belonging to the common 



* Among the Branchipodidae the genus Polyarieynia alone possesses a 

 larger number of segments (namely 22), nineteen of which bear appen- 

 dages. 



t The numbers refer to Apus cancriformis. 



