494 MR. A. W. WATERS ON 



do tliey ever refei- to Pallas, although many of the species dealt 

 with had been mentioned by that author. This is certainly 

 difficult to understand, and perhaps would have been different 

 if Solander had lived to complete the work himself. 



Hincks considered that the species in Solander's genus had all 

 found places elsewhere, and that the genus of Solander had lapsed. 

 He therefore took Cellaria as the genus of Lamouroux, who vmder 

 Cellaria put Cellaria (as now understood) with the type C. sali- 

 cornia and also Ttihucellai^ia. Lamouroux says of all genera none 

 seems to contain as widely distinct species as this, and that 

 it seems to have been formed to contain everything that could 

 not be placed under Flustra or Sertularia. 



Instead of dropping Celhdaria as hopeless, Busk unfortunately 

 retained it for a small division, and this has been accepted. 



The first species mentioned by Pallas in his genus is tuhu- 

 cellaria, which also is included in Lamouroux's Cellaria, though 

 not as the type, which was Cellaria salicornia; and Stoliczka*, in 

 a long discussion of the subject, took the view that as tubucellaria 

 Avas first mentioned the genus must be called Cellaria, and 

 Cellaria, as now understood, must be separated as Salicornia. 



We now see that it is unfortunate that Hincks should have 

 retained Cellaria, though it then appeared that this would not be 

 challenged, and it has been adopted generally, and no genus 

 seemed more firmly established. However, it is now clear that if 

 Hincks had continued to use Salicornaria there would have been 

 no possibility of the name of a now long recognised genus being 

 replaced by one used in most various ways, and now limited to 

 another small group. 



Those who are at work upon the class know how often the 

 descriptions of the earlier authors are now meaningless, for the 

 characters then used are found to be useless ; but this can hardly 

 be appreciated by those who have not had occasion to consult 

 such descriptions. It is as if some well known tree had genera- 

 tions ago received a name and a few lines of description which 

 would apply to a quarter of our phanerogams and some cryptogams. 



I have previously shown that we are brought into a perfectly 

 ridiculous position by being asked to adopt such names at all 

 costs, when we often have no idea what they meant. It is not 

 science, and since Cellaria as modified by Hincks is well established 

 I shall still use it f. 



* Foss. Bry. aus dem tert. Griinsand. der Orakei Bay bei Auckland, pp. 142-149 

 (1864). 



t Norman (" Polyzoa of Madeira," Journ. Linn. Soc, Zool. vol. xxx. p. 293 

 (1909)) challeng:es tlie correctness of considering that Tubucellaria opuntioides 

 should have been considered the type of Celhdaria, but I cannot agree with his 

 conclusions as to what I say being contrary to the British Association Rules of 

 Nomenclature. My edition is later than Norman's, but apparently is only a reprint, 

 and it says " When they omit doing so " (i. e. fixing a type), " it may still in many 

 cases be correctly inferi'ed that the first species mentioned on their list, if found 

 accurately to agree with their definition, was regarded by them as the type." This 

 rule of course means that if the generic diagnosis is taken from some one species, 

 and that an error has been made in including the first, then common sense may be 



