Classification of the Arthropoda. 371 



the normal form of somatic appendages than is observed in 

 regard to the second pair. I do not think it improbable that 

 at some future date Professor Claus may adopt the view which 

 I have advocated as to the first, just as he lias adopted it in 

 regard to the second pair of Crustacean antenme; and I am 

 therefore anxious to take the present opportunity of insisting 

 upon an important piece of evidence in its favour which has 

 come to light through my researches on the relationship of 

 Limulus to the Arachnida. Packard, as is well known, dis- 

 covered the " brick-red glands " of Limulus, the structure of 

 which I have since investigated (Quart. Journ. Micr. Sci. 

 January 1884). These glands are similar in essential 

 structure to the "shell-gland" of the Entomostracous Crus- 

 tacea. I discovered that they exist in Scorpio and also in 

 Mygah in a highly developed condition, and have given to 

 them the name " coxal glands," on account of tlieir relation 

 to the coxte of the prosomatic appendages. In none of the 

 Arachnids [Limulus, Scorjyio, and Mygah) do these glands 

 open to the exterior in the adult animal. But Mr. Gulland, 

 in my laboratory in London, and Mr. Kingsley, in Boston, 

 Mass., have independently ascertained that in the young 

 Limulus the coxal gland opens to the exterior on thehasal joint 

 of the fifth jpair of appendages [QwdiXt. Journ. Micr. Sci. 1885). 

 Now in the Crustacea Entomostraca the shell-gland opens to 

 the exterior at the base of the second pair of maxillse. If 

 we reckon the first pair of Crustacean autenna3 as the 

 equivalent of the first pair of appendages of the Arachnida, 

 as is the case according to my long since published view 

 of their nature, then we arrive at the striking result, , 

 pointed out by Kingsley, that the Crustacean shell-gland 

 and the Arachnidan coxal gland open in loth cases at 

 the base of the fifth pair of appendages. On the other hand, 

 if Professor Claus is right in considering the first pair of 

 Crustacean antennae as essentially prostomial, and in regarding 

 the first pair of Arachnidan appendages as the equivalent of 

 the second pair of Crustacean antennge, then the shell-gland 

 of Entomostraca loses its agreement in position with the 

 coxal glands of Arachnida, and has to be assigned to the 

 fourth pair of true somatic appendages instead of the fifth. 

 The argument is, I admit, not a conclusive one, since the 

 Pro-Arthropod must have been, like Peripatus, provided with 

 a nephridium (from which shell-gland and coxal gland are 

 derived) at the base of each pair of appendages. Never- 

 theless it has weight in a question which can only be decided 

 by the accumulation of converging evidence ; and it is, cceteris 

 paribus, more likely that the coxal glands and the shell- 



