GENERAL CONSIDERATIONS. 357 



lying between the ventral cord and the intestine, which affords 

 attachment to numerous groups of muscles. The views of Strauss- 

 DuRKHEiM were supported on embryological grounds by Ed. van 

 Beneden (No. 8) in 1871, and J. Barrois in 1878. Glaus also, 

 in 1876,* expressed his belief that "the air-breathing Arachnida 

 may have been derived from the polygnathan Merostomata (Trilobites, 

 Eurypterida, and Xiphosura)." Huxley had also expressed himself 

 in a similar way as to the genealogical connection between the 

 Arachnida and the Merostomata. The near relationship of the two 

 groups has been recently demonstrated in more detail by Eay 

 Lankester (No. 16), by means of a careful comparison of the 

 structure of Limulus with that of Scorpio. Although Ray Lankester, 

 as we think, goes decidedly too far in insisting that Limulus should 

 be regarded as an Arachnid, he still deserves credit for having 

 established on a broader base the view that the two forms belong 

 to the same phylogenetic series. It appears to us that the structure 

 of the limbs used for respiration and adapted to aquatic life, the 

 absence of the jNIalpighian vessels, and, moreover, their connection 

 with the Trilobites, which are further removed from the Arachnida, 

 afford sufficient cause for giving the Xiphosura a more independent 

 position. 



We cannot here vindertake to enter further upon the palaeonto- 

 logical evidence in favour of the genealogical connection between 

 the Arachnida and the Palaeostraca. We can only mention that 

 among the Gigantostraca, which are nearly related to the Xiphosura, 

 forms are found which, in appearance and in the segmentation of 

 the posterior region of the body, stand still nearer to the Scorpiones 

 than does Limulus itself. We must confine ourselves to a short 

 consideration of the points of comparison between Limtdus and the 

 Scorpiones. 



In both forms we recognise an anterior region of the body 

 (cephalo-thorax) carrying six pairs of limbs, and covered by a dorsal 

 shield, having on its upper side two median eyes, and, nearer the 

 edge, paired lateral eyes. The median eyes of Limulus and of 

 Scorpio agree so closely in structure, that we cannot doubt that 

 they are homologous. We might take the same view of the lateral 

 eyes, even though the numerous unicorneal lateral eyes of Scorpio 

 essentially differ from the remarkable lateral eyes of Limulus, the 

 composition of which is very primitive. In this case we should 

 have to regard the lateral eyes of Scorpio as of a modified type. 



* Untcrs. zur Genecd. Grundl. dcr C'rustac. Systems. 



