THE CEYLON rEARL-OYSTER. 319 



pai'ticiilar sections that I examined, there was a marked absence 

 even of the repair-substances. But in the zone immediately 

 around the point of contact between nucleus and shell, where, 

 when the nucleus was introduced, the epithelium of the mantle 

 was presumably unable to fit closely against the sui'faces, it is 

 otherwise. Here, on examining the whole '• Pearl " as a trans- 

 parent object after decalcification, an op;ique ring or zone was 

 distinctly visible. This was due to the presence of granular 

 matter, perhaps derived from the exterior, perhaps from the 

 tissues of the animal (text-figs. 39 & 40 B, gr.), and to very 

 irregular columnar and amorphous repair-substance (col., col.'). 

 This columnar substance is shown in greater detail in text-fig. 40, 

 A & B, corresponding respectively to col. and col.' in text-fig. 39. 



In text-fig. 40, A, in the niche between the nucleus and the shell, 

 where the mantle-epithelium could not reach, we see the product 

 of its secretion consolidated awaj^ from the influence of the 

 epithelium. Here the columns, instead of forming the character- 

 istic palisade-like structure, with their long axes perpendicular 

 to the secreting surface, are arranged in groups suggesting 

 incomplete spha;rocrystals. 



The curious fan-like arrangement which the columns take on in 

 text-fig. 40, B, suggests that the repair-substance ai-ose through an 

 extravasation of the shell-forming fluids at the point x, the layer 

 nac' representing the first normal nacre, secreted by the mantle 

 when it occupied that position, the irregular colvimnar and 

 amorphous substance being due to the consolidation of tlie 

 secretion which filled the space, triangular in section, which lay 

 between the nucleus (nu.), the shell [nac), and the mantle ; the 

 position occupied by the last named being represented by the layer 

 of nacre marked nac '. 



(13) Varieties of Ceylon Pearls. 



Pending a classification based on the nature of the causes 

 which give rise to the formation of the pearl-sac, I propose, 

 following Herdman, to sejiarate the pearls which I have examined 

 in or from the Ceylon Peai-1-Oyster into two main groups, 

 accordingly as they typically occur clustered in the neighbourhoods 



really satisfactory proof has ever been given that free spherical " pearls " can be 

 produced in this waj-, though Prof. Mitsukuri (/. c.) says that there are some hopes 

 that this will be done. There is no theoretical reason why a modification of the 

 Japanese or Linnuean operation should not be devised which would achieve this end- 

 indeed, there is some reason to think that Linnajus actually did produce some round 

 " pearls " and not only " blisters." But such bodies, if produced, would not \)e 

 " pearls " in the strict biological sense, though it is quite likely that they would be 

 marketed as such in quantities before the difference was detected. 



Since writing the above, I have been informed by Mr. Toyozo Kobayashi, Professor 

 at the Tokyo Higher Technological College, who is associated with Mr. Mikimoto 

 in his enterprise, that perfectly free '" pearls " have been produced by these methods 

 within the last two years, but so far only exceptionally, and on a scale so small as 

 not to be applicable connnercially. 



