20 Alexander Petntiikevitch, 



were in realit\- overlapping giil-j)lates which are shghtly longer than 

 each segment and are connected with the body wall at the anterior 

 edge of each segment. At least on the longitudinal section of one 

 specimen of Eurypterus remipes one can see on the outside the heavier 

 outline of the gill-plate with the mould of the gill-book and at a little 

 distance from it inward a thin black line representing evidently the 

 body wall. The appearance of this line suggests that the ventral 

 body wall of the mesosoma in eurypterids was not chitinizea, but 

 thin and soft. Herein it resembles the ventral body wall of Palaeo- 

 zoic scorpions, which as I thy to show in the Special Part was most 

 likely thin and soft. But a derivation of they scorion's lung-books 

 from such gill-books as those in Eurypterus seems highly improbable. 

 The line of attachment of the gill-plates in Eurypterus, as I have 

 mentioned above, is at the anterior edge of the somite, whereas the 

 spiracles of the scorpion are in the posterior half of the segment 

 Before an insinking could have taken place, the gill-plate must have 

 first moved down along the surface of the somite, a condition not 

 known to occur in any of the described eurypterids. Another blow 

 to a theory of the origin of scorpions from eurypteiids comes through 

 the beautiful work of Clarke and Ruedemann on the Euryptenda 

 of New York which was published while my present monograph 

 was still in the printing. In this exhaustive study of the rich material 

 obtained from different horizons the authors not only bring together 

 all that was known alread}' about eurypterids, but widen and deepen our 

 knowledge of the group in many respects. Although they seem to 

 be in error when they refuse to homologize the sternum of the scor- 

 pion with the metastoma of the eurypterids and the chilaria of 

 Limulus, they show conclusively that eurypterids have in some 

 respects more similarities with limuloids than with scorpions. The 

 presence of five pairs of gills, the large operculum of the second ab 

 dominal somite, the structure of the eyes, the large carapace and 

 smaller number of segments in embryonic stages are of especial im 

 portance. The long prosoma and thin metasoma of scorpion embryos 

 are the most important characters separating them from eurypterids. 

 On the other hand the fusion of abdominal segments even in the 

 oldest limuloid, Protolimulus eriensis, from the Devonian and the 

 absence of even partial fusion in eurypterids from the Cambrian to 

 the Permian suggests "that the hmuhds and eurypterids were prob- 

 ably separate in Precambric time." It is interesting to mention 

 in this connection that according to our authors the eurypterids of 

 New York may be divided into four groups in regard to their mode 

 of life and that the different deposits in which they are found show 



