COLLECTED BY DR WILLEY. 127 



The one figured is one of the largest observed, being two-thirds of the length of 

 the ovoid nematocysts. The larger examples of these three-spined nematocysts are 

 of comparatively rare occurrence, only a few being present in some of the tentacles 

 and being more commonly present in the tentacles of the gastrozooids. The smaller 

 nematocysts of this form have not more than one-sixth the length of the large ones." 



All of these points I have repeatedly observed. The large nematocysts in the 

 sections of some of Dr Willey's specimens exhibit a condition which we may assume 

 is one of incomplete extension, the vesicle containing a fibril spirally wound at the 

 base, which is continuous with the spiral fibril running in the lumen of the thread 

 (PI. XVI. Fig. 2). All of these nematocysts are fixed in the ectoderm of the 

 coenosarc. 



On examining the surface of the corallum with a lens or in searching through 

 the de'bris at the bottom of the bottle in which it was present, a certain amount 

 of white fluffy material may be found which consists of a felt-work of fully exploded 

 nematocysts. An examination of these with the microscope shows details similar to 

 those described and figured by Moseley (PI. XVI. Fig. 1, copied with slight modifi- 

 cations from Moseley), no spiral fibril in the vesicle and thread being visible with 

 ordinary powers of the microscope. Now it must be noticed that in the former the 

 nematocysts are in contact with living cells, and may themselves be considered to 

 be alive. In the latter the nematocysts may have been shot and dead sometime 

 before the specimen was preserved. 



It might be assumed that the fibril which is shown in Figure 2, drawn from a 

 nematocyst fixed in the ectoderm, is the terminal portion of the thread fixed in the 

 act of passing along the lumen of the basal portion of the thread to complete the 

 fully exploded nematocyst. Its form indicates however that it is much more plastic 

 than the outer wall of the thread, and as it stains deeply in haematoxylin, which 

 the wall of the thread does not do, it must have at least a slightly different 

 chemical structure. Moreover the spiral axis is equally present in fully exploded 

 threads (see below). The appearance that the fibril has in the thread is strikingly 

 like that of the myophan thread in the stalk of Vorticella, as Dr Willey remarks. 

 These notes on the structure of the nematocyst of Millepora have some significance 

 when taken in conjunction with the statement made by Dr Willey that the threads 

 are retractile. In a private letter to me he says, " I think there can be no doubt 

 now about the retractility of these threads. I did not observe it once or twice, but 

 repeatedly in different stocks of Millepora. The retraction takes place suddenly just 

 like a Vorticella." The retraction of these threads was observed with a lens, and it 

 was particularly noticed that they were confined to the coenosarc and had no relation 

 to the dactylopores or gastropores. 



These observations of Dr Willey's seemed to me of such importance that I felt 

 it to be essential to examine again with very great care the thread of the fully 

 exploded nematocyst to see if it exhibited any traces of the spiral fibril. For this 

 purpose I collected some of the felt-like substances from the surface of a colony, 

 stained it deeply in Kleinenberg's haematoxylin, and examined teased preparations of 

 it with Zeiss's 2 mm. oil immersion lens. 



