282 ME. H. M. BEENABD ON THE EELATION 



line between these anterior and posterior pairs of limbs is the 

 true waist or division between tborax and abdomen, cannot be 

 maintained. The fact that Ah/cus roseus has a waist, as de- 

 scribed by Kramer, may be due to the comparatively great length 

 of its abdomen, all the larger species beiag capable of bending 

 the body somewhat. It is, however, right to add that the claim 

 that the line dividing the limbs into two pairs is the true demar- 

 cation between thorax and abdomen does not rest "only on the 

 arrangement of these limbs, but also on the fact that, in some 

 Acarids, the mouth-parts are so specialized and altered from the 

 primitive Arachnidan type that some would^find in them 2nd 

 maxillae and an underlip presumably homologous with these same 

 parts in the Hexapoda. I shall return to this presently. 



My argument, however, does not rest only on the segmenta- 

 tion, although this is certainly the most important point. This 

 arrested development theory throws some light on the nature 

 and position of some of the internal organs. 



Pig. 3 B (PI. XX.) is adapted from Winkler's account of the 

 anatomy of Gamasusfucoruon. It shows the extraordinary position 

 of the heart. If, however, we compare this with fig. 3 A, we have 

 this position and shape very clearly explained. The Gamasus 

 heart is, as it were, only the first chamber of the adult Arachnidan 

 heart. The arrest of the further development of the abdomen 

 naturally arrested the further growth of the heart. 



This theory is of course difi'erent from that which sees pro- 

 gressive stages of degeneration in the hearts of the Phalangidse, 

 Chernetidse, and Acaridse. It is not impossible that some of the 

 Acaridse are to be deduced from larval Phalangidse, but not as 

 a result of degeneration, but as a fixation of larval forms. The 

 heart of Gamasus is only quantitatively degenerated ; relatively 

 to the size of the animal it stands on the same level of organiza- 

 tion as that of the adult Araneid. 



Pig. 4 B is a diagram of the alimentary canal of a Gamasus, 

 also after Winkler. If we compare this with the digestive tract 

 of an adult Arachnid (e. g. Mygale, fig. 4 A), we see what is the 

 part actually missing. It is again clearly the abdominal part, 

 i. e. that portion of the mid-gut which lies between the thorax 

 and the rectal vesicle. This offers further support to our theory 

 that some seven segments between the 1st and 2nd abdominal 

 and the anal segments were not developed. Here again we find 

 the difference is more quantitative than qualitative. 



