West-Indian Species of Madrepora. 23 



difficulty, though fragments may sometimes be indeterminable. 

 But the two forms are sufficiently allied for their separation 

 to be a mere matter of convenience. 



The difference between M. palmata and M. cervicornis is, 

 however, far more definite, and Pourtales had no hesitation 

 in keeping them distinct. Brook *, however, unites them ou 

 the ground that " Pourtales has pointed out, with regard to 

 the West-Indian specimens of palmata, cervicornis, and pro- 

 lifera, that the proper habit and robustness of each form is 

 associated -with a different position on the reef. M. palmata 

 grows in situations exposed to the force of the sea ; M. cervi- 

 cornis in less exposed localities; while for its full develop- 

 ment M. prolifera appears to require sheltered spots on the 

 inner side of the reef.'" This passage involves another 

 unfortunate misrepresentation of Pourtales, for, according to 

 that author, the species which " requires a rather sheltered 

 position for its full development " f is M. cervicornis, and not 

 M. prolifera. In respect to the position of growth of the 

 latter, Pourtales gives no information ; and, so far as my own 

 observations go, M. prolifera does not flourish in sheltered 

 spots inside the reef, but in deeper water than M. cervicornis, 

 and often outside the main reef. Thus the typical species in 

 the quiet coves of Parham Sound, Antigua, are M. palmata 

 and M. cervicornis. The best specimens of M. prolifera that 

 I obtained from Antigua came from the depth of 3 fathoms 

 from an exposed position on the slopes of Sandy Island. 



Pourtales did say that M. palmata is characteristic of the 

 exposed positions on the outside reefs, a statement, however, 

 which is only true with one important limitation. The 

 particular form of M. palmata known as " the car of Neptune," 

 which has a massive corallum formed of thick lamellar ex- 

 pansions, is no doubt the typical form of Madrepora found in 

 exposed positions in the West-Indian reefs. The fragile 

 branched coralla of M. cervicornis and M. prolifera would be 

 shattered if struck by the full force of a breaker ; they accord- 

 ingly grow in protected situations or at a depth below the 

 limit of the surf. 



It is possible that it was the "Neptune's car" form of palmata 

 which Pourtales had in mind when writing the previously 

 quoted remark. His statement, so far as my observations go, is 

 not correct for M. palmata as a whole. The alciform variety 

 of palmata, which both Pourtales and Brook include in that 

 form, grows under identically the same conditions as M. cervi- 

 cornis. For example, I collected specimens of both forms 



* Brook, op. cit. p. 28. 

 t Pourtales, op. cit. p. 84. 



