578 
lime; in other places the layer of lime was broken off or dissolved in 
smaller pieces of lime and big grains. In general I found the greatest 
quantity of lime against the parts of the periostracum that were folded 
most. The rest of the outer surface of the shell nowhere showed even 
the smallest traces of a crust of lime, so that we cannot help connecting 
the formation of it with the regenerate. Rubbel and Rassbach do 
not mention this crust of lime on the periostracum of the regenerate. 
This is perhaps owing to the fact that these examiners usually worked 
with shells into which no sand could penetrate, because of the paper 
coverings of the pieces of shell that were sawed out. The formation of 
the crust of lime might be caused by the stimulating influence effected 
by the grains of sand on the mantle, immediately after the shell of the 
mussel was damaged. This crust of lime must have arisen even before 
the formation of the periostracum of the regenerate. Then the complete 
regenerate ought to consist of 4 layers, the layer of crust of lime, the 
periostracum, the prismatic layer, and the layer of mother-of-pearl. How 
far this idea is correct, can only be proved by further examination of 
shells which are regenerated in a quite natural way. These are un- 
fortunately only to be found in about 1% of the cases, as I said already 
above. So it is easier to do the experiments of Rubbel and Rassbach 
over again, but then without any covering of paper etc. of the parts of 
the shell that are sawed out. Then they might also succeed in showing 
the layer of crust of lime on the periostracum. At the same time we 
might follow the influence of the water-temperature, the light and the 
quantity of lime in the water on the process of regeneration. As to the 
latter Villepoix found that in lime-free water, the regenerates con- 
sisted of periostracum only. None of the regenerates that Rubbel 
found in 1911 were connected with the edge of the mantle, through 
which the periostracum is made, according to the general opinion. 
Exactly what I found with my specimens in 1909. 
»Sie (the regenerates) sind vielmehr von dem unter den verletzten 
Schalenstellen gelegenen Mantelepithel secerniert. Es sind also die 
AuBenepithelzellen des Mantels fähig, außer dem Perlmutter, auch 
Periostracum und Prismenschicht zu bilden.« Thus Rubbel concludes 
and in 1912 Rassbach comes to the same results. I quite agree with 
this and so I must contradict what is said on page 684 of the 2"¢ edition 
vol. I 1910 of Parker and Haswell’s Text-Book: ”The periostracum 
and the prismatic layer are secreted from the edge of the mantle only; 
the pearly layer from the whole of its outer surface.“ In other books 
too, in Lang a. o., we find the same assertion; this therefore must 
be wrong. 
Leiden, 13. March 1912. 
