TUNICATA. 329 



my type-genus Polystyela, Giard, and naming the new family in strict accordance 

 with the rules of nomenclature. After the definition I added : " I form this family 

 for a very interesting little group of Ascidians, the position of which is difficult to 

 determine. I regard them as Compound Ascidians which are allied to the Cynthiidse 

 amongst Simple Ascidians, and have heen evolved from the suh-family Styelime." I 

 then gave an outline of the history of the genera which I considered would find their 

 place along with Polystyela in the family. The family was properly constituted in 

 1886, and the name cannot be altered until the name of the type-genus (Polystyela) 

 is changed. If Michaelsen can prove that Lesson's " Polyzoa opuntia" is the same 

 as Giard's Polystyela lemirri* then Polystyela will become a synonym of Polyzoa 

 and the family name will change to Polyzoidas, or, if it be regarded as a sub-family, 

 to Polyzoinae but not till then. In the meantime, if it -is placed as a sub-family of 

 Cynthiidee the name of the sub-family must be Polystyelinse. 



I may add that in the ' International Rules for Nomenclature,' now generally 

 recognised and followed, there is no direction that in forming a new family the oldest 

 generic title within the bounding line is to be selected as the type and give its name 

 to the family. The oldest genus may be quite unsuitable for such a purpose as it 

 may be an aberrant form very far from typical of the family. Surely it is only 

 common sense that it should be left to the founder of a new family to choose as the 

 type-genus that central form or assemblage of species which seems to him best to 

 typify the new group which he is defining. 



So far as to the family name now let me add a few remarks as to some generic 

 designations formed recently. 



Michaelsen, in 1900, expressed his dissatisfaction with the definitions of the 

 existing genera in this family, and introduced a new generic term, " Alloeocarpa," of 

 a provisional nature (lie saysf : " Dieser neue Name kann nur als provisorisch 

 angesehen werden ") for those species which have a certain character of reproductive 

 organs. But he adds: "Als Typus der Gattung Alloeocarpa, Mchlsn., mag 

 A. incrustans (Herdman) (= Synstyela incrustans, Herdman) gelten." He does not 

 sub-divide the old genus and does not retain any portion under the old name. He 

 merely substitutes a new name because he wishes to emphasize a new character. 

 Surely a better course would have been, if he finds that my Synstyela incrustans will 

 serve as a type of what he desires to put forward, to add the new characters (if 

 necessary) to the definition of the genus, retaining the old name Synstyela. As a 

 matter of fact, the unisexual character of the polycarps was described and figured in 

 the case of Synstyela incrustans in the '"Challenger" Report' (1886), and in all 

 probability holds for other species of Synstyela. It would be simple to restrict the 



* Michaelsen has suggested (' Mitteil. Naturh. Mus. Hamburg,' xxi., 1904) that Giard's genus 

 possibly does not belong to the Polystyelida?, but I find no basis in fact for this idea. 



t "Die holosomen Aseidien des magalhaensisch-siidgeorgischen Gebietes"; in ' Zoologiea,' Bd. xii., 

 Heft 31 ; Stuttgart, 1900. 



2 U 



