thfi Enrfosternite o/" Scorpio. 23 



It seems to me that, in order to answer tlie question, " Is 

 Limxihis an Araclinitl ? " we ought to compare Limulus with 

 GaJeodes ! Indeed, I must liere express a surprise, whicli 1 

 have long felt, that any serious attempt should have been 

 made to establish relationsliip between two groups by com- 

 paring their specialized forms, practically ignoring the 

 remaining representatives of the groups. And yet this is 

 what has been attempted. The Merostomata and the 

 Arachnida are supposed to be related because of the likeness 

 between certain Eurypterids and Scorpio. Scorpio was 

 clearly an Arachnid, and therefore related to other Arachnids, 

 and Linmhis was equally clearly related to the Trilobitcs. 

 To attempt to build up an elaborate proof that Limuhis must 

 be related to Scorpio on the ground of their external and 

 internal resemblance, while Galeodes on the one hand and the 

 Trilobites on the other are but incidentally considered, can 

 only lead to confusion*. The only scientific method of 

 dealing with the question of the alleged relationship between 

 Limulus and Scorpio lies first of all in the determination by 

 comparative morphology of what is the essential diagnosis of 

 the Arachnids and what is that of the Merostomata f. 



Having seen that the endosternites of the Arachnids are 

 apodematous structures due to fusion and compression of 



* Galeodes, indeed, is often not even considered at all ! 



t I would here point out tli.at it is still a matter of dispute whether 

 the PycnogonidfP and Pentastomidae are or are not Arachnids. It seems 

 to me that the only way to obtain a true insight into the essential mor- 

 phology of the various groups of the Artie ulata is, by comparing all 

 the available forms, to discover how the original undifferentiated 

 condition of the segmentation of the ancestral fonn has been modified in 

 each group. I have already endeavoured to do this for the Crustacea, 

 with results which are in a way to being more rapidly confirmed than I 

 ever expected. I am now engaged on a comparative study of the 

 Arachnids, and I hope to find the clue to the original modification of their 

 primitive se^'nientation. And here I should like to add that, although 

 my friend Mr. Pocock's paper, " On some Points in the 3Iorphology of 

 the Arachnida '•' (Ann. & Mag. Nat. Hist., Jan. 1893) is of much service 

 as a store of facts, the arguments which he bases upon them are not con- 

 clusive. He endeavours to show tliat all Arachnids mic/ht be deduced 

 from a scorpion-like ancestral form. I have recently shown in ' Nature ' 

 (Nov. 16, 1893) that there is no necessity to assume the strict homology 

 of the stigmatic segments, on which assumption much of Mr. Pocock's 

 argument appears to rest, iuasmucli as all the segments originally 

 possessed tracheal invaginations. Further, it is not sufficient to take 

 simply the nitmher of the segments into account ; the fusioiis of the 

 segments are, if anything, even more important, those animals with the 

 lesser number of fused segments being as a rule more primitive than those 

 with a greater number. Judged by this standard, Galeodes or Schizo- 

 nohis, and not Scorpio, is the primitive Arachnid. 



