356 NATURAL SCIENCE. May. 



side ; but in some of the Buthidae there are as many as five, and the 

 fact that these five may be observed in genera that are not especially 

 related points to the conclusion that the ancestors of this family 

 possessed this number ; and this conclusion seems to be borne out by 

 the circumstance that the nearest living ally of the Scorpions, namely, 

 Thelyphomis, which we believe to be the most primitive of all the 

 Epectinate Arachnida,' may also present five lateral ocelli on each 

 side. 



The conclusion that the Buthidae are, on the whole, the most 

 archaic of living Scorpions is borne out b}'' a study of the geographical 

 distribution of these animals ; for, as will be seen in the sequel, this 

 family is represented by peculiar genera in all the quarters of the 

 globe where Scorpions exist. 



The absence of proofs of their great antiquity, which was referred 

 to in connection with the inter-relationships of the existing species of 

 Scorpions, is also revealed no less remarkably by a study of their 

 geographical distribution ; for in this latter respect they show con- 

 siderable resemblance to such a relativel}' recent group as the 

 Mammalia. No doubt our knowledge of the recent species of this 

 latter class is very much more complete than is our knowledge of the 

 Arthropods in question ; but so far as we have been able to ascertain, 

 Wallace's map of the Zoological Regions of the world, if we eliminate 

 the northern parts of the so-called Palaearctic and Nearctic Regions 

 where Scorpions do not occur, is closely applicable to this group of 

 animals. We may, therefore, without further words of preface, 

 proceed to discuss the characteristic features of the various regions. 



In the Palaearctic Region the northern limit of Scorpions in 

 Europe and Western Asia appears in a general way to be defined by 

 the mountain-chain of which the Pyrenees is the western and the 

 Caucasus the eastern extremity. East of the Caspian they have not 

 been recorded north of the 40th parallel, but they are known to occur 

 at two places upon this line, namely, at Kashgar in Turkestan, and 

 at Pekin in China. The two species from these localities are 

 nearly allied to each other, and they unmistakably belong to the 

 Mediterranean section of the genus Buthus. Furthermore, a species of 

 Btithtis, with distinct Palaearctic affinities, has recently been recorded 

 from Shanghai. Consequently, w^e must conclude that the Medi- 

 terranean fauna extends from Spain and Algeria to the coast of 

 China, with the probable omission of the table-land of Tibet. The 

 southern limit of this area in China is, for want of data, uncertain ; 

 but since the Scorpions of Burma, Assam, and North India are totally 

 different from those of Turkestan, Afghanistan, and Persia, we may 



1 Since, as Mr. Laurie has pointed out, the term Ctenophora, which I have 

 recently assigned to the Scorpions as opposed to all other Arachnida which were 

 called Lipoctena, is generally recognised as signifying a group of Coelenterata, 

 I wish to propose the terms Pectinifera and Epectinata as substitutes for the 

 above. 



