42G GENERAL CONSIDERATIONS ON 



an extraordinarily early period before the germ-band has appeared 

 {blastodermic cuticle of the Crustacea), or else before the limbs have 

 formed {embryonic cuticular envelope of Limulus, deutovum-membrane 

 of the Acarina, embryonic envelopes of Pentastomum). All these 

 cuticular envelopes then form a further covering to the embryo 

 within the egg-integument. 



Zoologists were for a long time inclined to ascribe to the larvae of 

 the Arthropoda an important phylogenetic significance. But when it 

 was recognised that these larvae often represented secondarily modi- 

 fied (adapted) forms (Nauplius, Zoaea, Pantopodan larva, cater- 

 pillar of Insects), the comparison of the adult forms received more 

 attention, a far higher value being set upon this branch of inquiry. 

 The recent advance in the knowledge of Peripatus has been of special 

 significance in interpreting the Arthropoda, and in tracing them, back 

 to lower forms. Too great importance was indeed attached to those 

 characters of Peripatus that pointed to the Annelida, and gave rise to 

 doubt as to the uniformity of the Arthropod stock.* As it was seen 

 that the Myriopoda and the Insecta could be linked on directly to 

 the Annelida through Peripatus, the only way out of the difficulty 

 caused by the Crustacea, which were apparently far removed from 

 Peripatus and in some respects showed less primitive conditions, 

 was to assume for them an independent origin for the Annelidan 

 stock. Recent research has, however, made it appear that Peripahis 

 is more closely related to the Arthropoda than was formerly assumed. 

 The nephridia are closed by end-sacs (remains of the coelom), and 

 show the type which we find recurring in the antennal and shell- 

 glands of the Crustacea. The permanent body-cavity is a pseudocoele 

 which develops after the disintegration of the coelomic sacs through 

 the enlargement of the primary body-cavity. The heart is of the 



* Doubts of this kind have repeatedly been expressed. They have found an 

 able exponent in the anonymous author of an article in Kosmos (No. 1), who 

 argues against the unity of the Arthropodan stock. Oudemaxs, in the same 

 way, is in favour of breaking up the division of the "so-called Arthropoda" 

 (No. 8), and Fernald, in his recent treatise on the "Relationships of the 

 Arthropoda" (No. 4), gives indications of holding a similar view. This latter 

 author, indeed, derives the three great principal trunks of the Arthropodan 

 stock, the Crustacea, Arachnida, and Insecta, from a common root. This root, 

 however, does not spring from the Annelida, but reaches back to the unsegmented 

 forms from which also the Annelida are derived, though in another direction. 

 Peripatus then branches off, and thus is not directly connected with the three 

 great branches of the Arthropodan stock, the Myriopoda also being independent 

 of these. These latter are, however, thought by Fernald to be connected with 

 /'< ripatus. In any case the complete uniformity of the Arthropodan stock is not 

 held by these authors, and it is also opposed by KlNGSLEY (No. 7), who, in 

 spite of their many points of agreement, derives the Crustacea and the Insecta 

 from different starting-points. [See also No. XII. — Ed.] 



