78 



Fseroes is that it cannot be regarded as a typical repre- 

 sentative of the species to which it seems most reasonably 

 referable, namely Parantipathes larir. It diti'ers from the 

 type in having several branches from the main stem, in 

 not having the pinnules of a row strictly in one plane, in 

 having more numerous pinnules (over 20 in three cm. 

 of one vertical row, instead of about 11), and in hav- 

 ing flexible, not " rigid," pinnules. But in Brook's detailed 

 description it is stated that the pinnules of a row are 

 "almost in one plane," and we may also note that the 

 pinnules of a fine specimen of Parantipathes larix from 

 Naples, are anything but rigid in the sense that an icicle 

 is rigid. They are flexible like a foil. 



Our specimen agrees with Parantipathes larix in having 

 the pinnules in six vertical rows (though these are decidedly 

 irregular), in having six pinnules to one revolution of the 

 axis, in showing three longitudinal rows of spines on one 

 aspect of a pinnule, and in having similar spines not 

 uniform in size. "We have compared a pinnule of a Naples 

 specimen with a pinnule from the Faeroes specimen, and 

 we find them practically identical. Therefore we conclude 

 that the Fieroes specimen is referable to P. lariv, although 

 it deviates from the type in certain respects, such as 

 branching. The deviations may be growth-reactions to a 

 very different environment. 



We make no apology for entering into details on this 

 subject of specific identification, for in all such questions 

 one is only too apt to err in the opposite extreme. At 

 first sight it seemed impossible to identity the weathered 

 specimen from the Faeroes with a perfectly preserved 

 specimen from Naples, yet we think that this is necessary. 

 This means that Brook's description of the species as («) 

 with stem "simple or rarely branched near apex," {b) 

 with pinnules in six planes, and (c) with pinnules rigid, 

 about 11 in each vertical row for 3 cm., requires to be 

 modified in regard to each of these points. When we 

 think of plants, e.(/., forms like " Shepherd's Purse " and 

 " Whitlow Grass," to which the ramose colonial Ccelentera 

 are in so many ways comparable, or when we think of 



