10 Alexander Petrunkevitch , 



a variable number of appendages from 4 pairs in Pycnogoniim littorale 

 to 8 pairs in Decalopoda. The order Tardigrada has also been con- 

 sidered to belong to the class Arachnida, as highly degenerate forms 

 descended from Acari. This view has recently been abandoned on 

 good grounds, since the internal organization of these little inverte- 

 brates has little in common with either typical or aberrant forms. 

 It has been pointed out by different authors that a complete absence 

 of a meso- and metasoma and the opening of the sexual ducts into 

 the intestinal canal, has no homologon among Arachnida. It is true 

 that the abdomen has almost completely disappeared in the case of 

 Pycnogonids also, and that the sexual ducts in this group also open 

 in a wa\' not similar to that in Arachnida. But the Pycnogonids 

 have, at least, segmented appendages, while the so called feet of the 

 Tardigrada are not segmented. But the absence of segmentation 

 may be due to the loss of it during the phylogenetic development just 

 as it has been lost in the abdomen of true spiders and ticks, where it 

 appears only in the embryonic development. Moreover, the legs of 

 some mites show a considerable tendency in the same direction of 

 reducing the normal number of segments and of obliterating the ex- 

 ternal signs of segmentation. The order of Pentastomida is also 

 frequently placed in the class of Arachnida. The four chitinous 

 hooks of the adult are homologized with the first and second pair 

 of appendages and derived from the two pairs of appendages of the 

 larval stage. The loss of the remaining four pairs is attributed to 

 the highly specialized parasitism of the Pentastomida. But if these 

 forms have e\'er developed from typical Arachnida, there is certainly 

 no evidence of their past history left either in the anatomy- or the 

 embryology of the now living forms. It may be safer, therefore, 

 to derive both the Tardigrada and the Pentastomida directly from 

 primitive, annulated worms, and to regard them as separate phyla. 

 In the internal organization of Arachnida there are several struc- 

 tures in common to the majority of them, such as the coxal glands 

 for example, but the influence of parasitism extends to internal 

 characters as well as to external structures and has resulted in a 

 degeneration of many organs. Thus the characteristic organs of 

 respiration are wholly lacking in many mites and in other forms have 

 followed two different courses of development The division of the 

 body into a ccphalothorax and an abdomen, also holds good only in 

 the case of typical Arachnida. In the Solifugae we find the ccphalo- 

 thorax not yet completely fused, since the last three segments are 

 free ; in the Ricinulei there is a movable plate in front of the first 

 pair of appendages ; in the Opiliones the bod\- shows a tendency 



