73 



tube near the middle of the back, and Professor Miiller in liis ' Sy- 

 stem ' uses the presence or absence of this tube, ■vvhich he regards 

 as a vent, as a character to separate the class into two divisions ; but 

 I think his table of genera sho\vs that this division can scarcely be 

 considered as natūrai, for he has been obliged to separate species of 

 Astropecten from their allies, and to place them, on this single cha- 

 racter, in another division of the family. Secondly, it is very diffi- 

 cult to observe the presence or absence of this part, especially in 

 Astropecten, on account of the paxilU, and some species, uhich are 

 said to be without it, may have it ; for it is to be observed, that M iiller 

 and Troschel place the genus in ■vvhich Sabine first observed the 

 vent, in the family characterized as not having one. 



I mušt consider their work as a retrograde movement, after the 

 publication of my paper, w'hich they quote ; for though they might 

 not adopt the genera, yet it cannot but be allowed, that what I have 

 considered as genera are natūrai groups ; and it would have facilitated 

 the making out of the species they have described, if they had used 

 them as sections ; they have done so in a few instances (thus after 

 the publication of their paper they have divided the genus Goniaster 

 into tvvo, adopting my sections as their genera ; but as in the case of 

 Asterias, because they have divided it, they blot the names from the 

 system) ; thus their first section of Ophidiaster is the šame as my 

 genus, and their second is my genus Linkia, and the second section 

 of Asterocanthion appears to be my Tonia. 



It has alwa)'s appeared to me, that the great advantage of dividing 

 the species into small groups (let us call them genera or sections, 

 as Mve may) is, that it enables one more accurately to determine and 

 neatly describe and distinguish the species, and prevents the neces- 

 sity in each description of repeating what has been given as the 

 character of the group, as is the case in the system of Star-fish. 



Lastly, I suspect that had M. Miiller had the opportunity of 

 esamining and comparing the number of specimens of this genus to 

 be found in English coUections, he woiild have come to the šame 

 conclusion as I have done with regard to the distinctness of several 

 species which in the work above referred to he has regarded as mere 

 synonyma of some well-known species. At the šame time it is re- 

 markable that it should not occur to M. MūUer, that ■vvhen the spe- 

 cimens on which a certain number of species have been established 

 are contained in a single coUection, and divided into minute groups, 

 and arranged side by side, it is not so easy to make mistakes in this 

 particular as when the materials are to be coUected from various 

 scattered museums ; as the differences and the similarities are then 

 more easily to be seen, and any errors which may have been made, 

 more easily discovered. 



Thus I am convinced, if he had seen the series of specimens of 

 Asterias Helianthus and Cumingii, and A. multiradiata, \vhich have 

 passed through my hands, and the selection of them in the Museum 

 collection, it is quite impossible that he could have confounded them 

 into a single species. The šame may be observed with regard to 

 Linkia Ty pus, L. Bromnii, L. bifasciatus and L. unifasciatus ; vvith 



