COLLECTED BY DR WILLEY. 125 



It will be noticed that at the summit of some of the tubercles there is a key- 

 hole aperture, considerably larger in diameter than the gastropores, which is surrounded 

 by one or in some cases two ill-defined circular grooves. These marks indicate the 

 presence of Pyrgoma milleporae just below the surface and there can be little doubt 

 that the irritation caused by the growth of the cirripede gave rise to hypertrophy 

 of the coenosarc and produced the tubercle. In many cases however tubercles are 

 present which have no cirripede holes at their summits, but the following considerations 

 prove that the tubercles are nevertheless the results of parasitic irritation. A number 

 of tubercles showing no external signs of the parasite were filed down to a level 

 with the surface of the corallum or deeper, and others were fractured vertically with 

 a bone forceps. In half of these an irregular cavity was found containing the remains 

 of cirripede shells ; in the remaining half no such cavity or remains were found. 



The apparent absence of the cirripede remains may be due to incomplete 

 examination, i.e. they may have been deeper down than I was able to get by filing 

 from the surface, or the parasite may have been completely destroyed by the parasitic 

 fungus Achhja, and the hole subsequently filled up with fresh corallum. The fact 

 that in a large number of cases the tubercles are associated or were associated with 

 parasites suggests that the tuberculate surface cannot be an intrinsic specific character, 

 and this is confirmed by the fact that in the W. -Indian Millepores and in the 

 Millepores from water 5 — 7 fathoms in depth parasitic cirripedes are rarely found and 

 the surface of these corals is nearly always smooth. 



Facies veri'ucosa. 



I have given a figure of the largest and finest specimen in Dr Willey's collection 

 as it forms a fitting termination to the series which has been described. If we were 

 only acquainted with Millepores similar to the one figured in Fig. 8 and the one 

 figured in Fig. 1, it would never have occurred to anyone that M. verrucosa and 

 M. dichotuma were not good species. Yet the evidence which has been accumulating 

 for the past ten years appears to me to prove that they ought not to be considered 

 even as " varieties " of the one species. I believe that sister larvae might produce 

 forms as widely different as these if they hapjiened to settle on parts of the reef 

 where the conditions of tide, bottom, food, parasites etc. were widely different. 



The particular specimen is characterised by a surface covered with numerous 

 tubercles of approximately the same size and shape, the key-hole cirripede aperture 

 may be seen on the summit of many of these tubercles but in the majority of 

 them the cirripede has been overwhelmed. The base of this specimen (not shown in 

 the figure) which was broken across exhibits splendidly the dead cirripede shells 

 corresponding with the tubercles. There is one more feature about this photograph 

 which calls for notice. A broad meshed network of dark lines may be seen covering 

 the corallum, running principally between the tubercles. These lines are marked by 

 the deposit of a yellowish pigment, and I believe they represent the course of the 

 principal and largest canals of the coenosarc such as those figured in Moseley's (i.) 

 Plate XIV. Fig. 4 C, and in my woodcut 1, in Q. J. M. S., Vol. xxxu. The 

 marking out of these canals is not easy to explain. I have only noticed it two or 



