1898.] SPECIES or mtllepoka. 251 



gastropores which I have given above was obtained from the base 

 of a massive specimen, while the smallest was obtained from a 

 growing-edge of a slender specimen. 



Moreover, it occurred to me that if the size of the gastropores 

 is dependent upon their age or the rate at which the gastrozooids 

 have grown, there ought to be, in some cases at any rate, a 

 difference between the average size of the gastropores on one side 

 of a branch or plate and that on the other ; those on the face most 

 favourable as regards food-supply in the living state should be 

 larger than those on the other." Measurements confirmed my 

 point, and I found a difference in two out of three specimens 

 between the gastropores on one side and those on the other as 

 great as 0-03 mm. 



(3) The Degree of Isolation of the Cycles. — Moseley noticed that 

 in one specimen of Millepore taken at Zamboauga the cycles were 

 much more distinct than in other specimens, and suggested that this 

 feature might be of specific value. After very careful consideration 

 I am convinced that it cannot be. In many large specimens it 

 will be seen that the cycles are much more distinct in one part 

 than another. Sometimes the cycles are so crowded as to be 

 indistinct at the edge, and perfectly clear on the face or at the 

 base. The evidence points to the conclusion that in slow-growing 

 Millepores in unfavourable situations the cycles are distinct, and 

 that in fast-growing specimens in good situations the polyps are 

 formed in such great numbers that the cycles become confused. 



(4) The Presence or Absence of Ampulla;. — The ampuUte of 

 Millepora were discovered by Quelch in a specimen obtained by 

 the ' Challenger.' He founded a new species for the specimen, 

 which he called M. murrayi, and used this feature as an important 

 specific character. 



I have found that ampullae occur in plicate, ramose, and digitate 

 specimens, and, as will be explained later, the absence of ampullae 

 in any particular specimen merely means that at the time it was 

 taken it was not in a state of sexual activity. 



It is greatly surprising how very rarely specimens are found in 

 this particular condition, but 1 believe that it must occm- in all 

 varieties at one time or another in their life-history. 



(5) The Texture of the Surface of the Corallum. — The species 

 M. verrucosa of Miliie-Edwards, M. tuberculata of Duchassaing, 

 and M. striata of Duchassaing and Michelotti have been named 

 after the peculiarities of their surface. 



I have had an opportunity of examining a very fine specimen of 

 a Millepore, resembling very closely the type of M. verrucosa, and 

 I found that on the summit of a very large number of the verrucas 

 there is a small hole of the shape of a keyhole, which leads into a 

 cavity formed by a parasitic cirripede (probably Pyrgoma milleporce). 

 On others, however, no such evidence of parasitic interference with 

 normal growth is apparent from the surface, but nevertheless 

 there is reason for believing that the tubercle may have been due 



