COMPAEATIVE MORPHOLOGY OF THE GALEODID^. 395 



some cases an area more or less marked to indicate their former presence. From the 

 character of the limljs which persist as such, we learn that they did not differ from those 

 of the cephalothorax, but were filamentous jointed appendaii:es like the walking-legs. 



It is practically impossible to obtain these conditions from Lhnulus or the Eurypterids, whose 

 abdominal limbs were too highly specialized ever to give rise to a series of filamentous rudiments such 

 as are found in the Arachuida, and found, moreover, specialized in such various ways and for such 

 diflerent functions that we are obliged to assume that they are all developments of some simple 

 undifferentiated form of limb. It has further to be remarked that the recent demonstration of the 

 existence of Phyllopodan limbs in the rudimentary segments of the Trilobites explains the lamellate 

 character of the abdominal limbs of Limulus. We have, then, to ascribe very different origins to the 

 abdominal limbs in Limuius and in the Arachnids; in the one they are primitive Phyllopodan limbs, 

 persisting in the interests of respiration, while in the Arachnids they are the remains of a series of 

 filamentous locomotory appendages. 



Until quite recent times, again, the rudiments of abdominal limbs in the Arachnida were thought to 

 be confined to the first six abdominal segments, that being the number and arrangement of abdominal 

 appendages in Limulus. But the traces of abdominal limbs in the Arachnida are not confined to these 

 six segments, for they occur in all the segments. 



IG. The Muscular Si/stem. — The primitive Arachnid possessed a typical Annelidan 

 musculature, specialized (1) for the movement of locomotory and seizing appendages ; 

 (2) anteriorly, in adaptation to the new positions of the anterior ap^^endages, thereby 

 leading to the formation of the cephalic lobes ; (3) posteriorly, in adaptation to the 

 differentiation of some or aU of the abdominal segments into a distensible sac ; and (4), 

 in those cases in \\ hich the terminal segments were sj^ecialized into a ' taU,' for the move- 

 ments of that appendage. In addition to these, we have the primitive oesophageal 

 muscles developed into the powerful expanders and contractors of the jiumjoing-apparatus. 

 A well-developed series of dorso-ventral muscles extended certainly as far back as to the 8th 

 abdominal segment {Theli/2}fionus},]yrohab\j still further. Besides, a study of the various 

 existing arrangements of this musculature in the Arachnids leads to the conclusion 

 that they can only be deduced from an original Annelidan condition, each along its 

 own lines. 



Inasmuch as the musculature of Limulus is also a sijecializatiou of a typical Annelidan musculature, 

 it is possible to homologize all those muscles which can be certainly shown to be derivatives of the 

 same Annelidan muscles. The original argument which Prof. Laukester based upon the muscular 

 system breaks down in so far as he confined himself to one Arachnid, Scorpio, which is not 

 primitive in the matter of its segmentation. We may add, further, that the muscular system of the 

 Arachnida could never have passed through anything like the extraordinary specialization of the 

 Limulus musculature, once again to revert to its primitive condition, to be again specialized into the 

 Arachnidan type. 



Further, as a great part of the ai-gument in favour of the relationship between Limulus and 

 Scurpio is based upon the great external resemblance between the two, when we consider how intimately 

 associated the muscular and skeletal systems are, we are unable to understand the divergence of their 

 muscular systems [cf. Beck and Benham). Lastly, while the musculature of Limulus can be shown to 

 be a further development of the simple musculature of Apus, which is itself a specialization of the 



52* 



