8 MADREPORARIA. 



under a heading " B. septis pallvlisque hirtis," whereas, according to my Paris notes, these 

 elements are quite smooth in Lamarck's original coral. 



Up to this date, then, we find a struggle between the two tendencies ; on the one hand, 

 a kind of instinct to believe that every form discovered must belong to one or other of " les 

 especes connues," although all that was really known was little more than a few book names * ; 

 and on the other, the conviction forced home by actual observation and comparison that the 

 differences were profound, and necessitated separate descriptions and names. Unfortunately, 

 from tliis time onwards, it was the former tendency which carried the day. 



In 1871, Pourtales studied the Florida reefs, a locality quite distinct from any of the 

 localities we have yet named, and from which, so far as we know, no Pontes had been 

 recorded, or if it had been, it had not found its way into the larger treatises here referred to. 

 Both encrusting and branching forms were found in great abundance. Without a word of 

 comment, the former were assumed to be of the same species as the one Lamarck called 

 astrcemdes, while the latter were divided between the supposed species clavaria and furcata. 

 This process is perfectly natural according to the belief that coral " species " have wide 

 indefinite distributions, and that, given any single specimen of coral, we may assume the 

 existence of more or less extensively distributed " species," as we certainly could do if the 

 CJorals were endowed with the same powers of locomotion as is possessed by bird and fish. 

 But we are beginning now to discover that this can no longer be assumed of the Corals. They 

 are, so far as we know, local forms. Consequently, we begin to realise that we shall never 

 recover Lamarck's forms clavaria and furcata until we discover the localities whence they were 

 originally brought. 



It is, of course, possible that they may have come from the Florida reefs, but it is far more 

 probable that they came from some West Indian island with which France was at that time 

 apparently in more frequent communication than with any other parts of the American seas. 

 What is certain is that the attempt to force all branching Porites into one of two purely 

 imaginary species " clavaria " and "furcata " began from this date. The wish referred to above 

 seems to have begun to assert itself. 



In the year 1880, the same writer f worked over the corals brought from the Florida reefs 

 by Agassiz, and wrote the explanation of the Plates. Three different branching forms are 

 illustrated, different not only in form of growth, but, if anything, even more strikingly different 

 ill the calicles. One is called clavaria, and the other two furcata. All three of them are very 

 different from Lamarck's originals. 



One almost instinctively asks what can be the difference between clavaria and furcata if 

 ordinary differences are so completely ignored. It must be sometliing very profound ! It is, 

 therefore, somewhat startling to find that it is nothing more than that clavaria is supposed to 

 be more club-shaped and furcata more openly forked. We are, consequently, not surprised to 



* And perhaps Milne-Edwards' beautiful, but generally neglected figures of P. furcata, see 

 above, p. 6, 



t Mem. Mus. Comp. Zool. pi. xii. figs. 4, 7 ; pi. xvi. 



