434 M. N. Cliolodkowsky on the 



exist, which, however, are perhaps more apparent than real. 

 Thus the last (tenth) abdominal segment of Machilis bears 

 three long-jointed appendages which are similar to one 

 another. If the two lateral appendages correspond to the 

 cerci of the Orthoptera, which is hardly open to doubt, what 

 is the value of the third median appendage ? Haase expresses 

 the conjecture that this represents nothing else than a much 

 elongated and secondarily jointed end-segment (anal covering- 

 piece). This explanation is a very plausible one. The best 

 proof of the iact that the segments of the Insect body may 

 exhibit secondary (superficial) segmentation is furnished by 

 certain larvae, such as, for instance, that of Chardio^liorus^ 

 which exhibits twenty-six apparent segments (behind the 

 head), that of Thereva, in which seventeen apparent abdo- 

 minal rings are visible, and others (according to Perris). 

 But it is self-evident that a definite judgment upon the 

 median terminal filament of Machth's, Ephemeridas, &c. 

 cannot be pronounced until the embryology of these forms 

 has been investigated. Another very interesting example is 

 furnished by the genus Tridactyhis^ Oliv. {Xya, Latr.), in 

 which the tenth abdominal segment bears two pairs of cerci 

 (ventral and dorsal). In this case the ventral cerci perhaps 

 correspond to the rudiments (which in other Insects, as also in 

 Phyllodromia J undergo degeneration) of the tenth embryonic 

 abdominal segment, which subsequently fuses with the 

 eleventh. This question has likewise to be decided by 

 embryological investigation. 



Thus we see that the difficulties as to the interpretation of 

 cerci, to which reference has been made, are at any rate 

 capable of more or less plausible explanations ; on the other 

 hand, the development of the cerci in Phyllodromia shows so 

 clearly that they are equivalent to the antennse and the rest 

 of the ventral extremities that I can discover no reason 

 whatever for not regarding them as homologous with the 

 thoracic limbs. In any case such an interpretation of the 

 cerci appears to me to have a much better foundation and to 

 be less arbitrary than, for instance, the comparison with the 

 furcal appendages of A2ms or the anal feelers of Polychates 

 (Haase). 



Of the highest interest arfe the ventral appendages of 

 Poduridge, that is to say the springing-fork and tlie so-called 

 ventral tube. Haase considers the fork to be equivalent to 

 the abdominal styles of MacMUs^ and therefore not homolo- 

 gous with the limbs. But it has already been shown by 

 Uljanin* that the springing-fork of thePoduridse arises from 

 * B. Uljanin, " Beobachtungen iiber die Entwicklung der Poduren," 



