90 
Proceedings of the Royal Society of Edinburgh. [Sess. 
amphibians replacing a lost leg. The typical decapods selected for the 
present work would be handicapped by such a limb, and the fact that they 
can lay a leg down in miniature and expand it only when it is practically 
ready for use must be a considerable advantage to them. The papilla, too, 
has very complete sensory innervation, and this must provide for its 
careful protection by the crab. It is devoid of the calcareous coating 
which covers other parts of the body, and so is able to increase in size, 
unlike the animal as a whole. Even as the form of this miniature limb 
suits the general requirements of these decapods, so is the form of the 
papilla itself modified to meet the needs of each species. The lobster in 
nature is practically always under water. When removed from its element 
its straight papillae droop and tumble about like paralysed limbs (fig. 16) ; 
below water they stand out under cover of the other legs and seem to be 
perfectly well supported by their own rigidity. The shore crab, on the 
other hand, is usually living in air half its time. In it the papilla is 
compactly rolled up within a tough envelope, and quite firm in air or 
water. The hermit crab is intermediate between the other two forms 
both in its life-habits and in the shape of its papillae. The perfection of 
papilla formation also varies directly as the frequency of injury. In the 
shore crab 50 per cent, of individuals are regenerating in the spring of 
the year, whereas in the lobster regeneration of limbs is only found to 
the extent of 2 per cent, or 3 per cent. 
The second point of note in the regenerative process is that it goes on 
at the expense of the body, and, as was the case with my experiments on 
Eupagurus, in the entire absence of food. This is very like what 
happens in the growth of a malignant tumour in the higher animals, and 
I have often seen the analogy carried so far that the crab dies when the 
process has been completed. This points to an influence in the cells them- 
selves towards multiplication, rather than to a local nutritional stimulus to 
division, as many have sought to prove. The experiments also show, as 
those of previous workers have shown, that the rate of regeneration 
decreases as the animal becomes older. In short, it is dependent on the 
frequency of moulting, which decreases with age, and it is most marked in 
the period leading up to moulting. 
Thirdly, the problems connected with the advent of function in the new 
muscles have yet to be attacked. It may be here noted, however, that 
though the fibrillse of the new fibre are anatomically complete, function 
does not come till several days after moulting has occurred. Its first 
appearance is seen in rhythmic movements of the limb, as if only a few of 
the stimuli per second passing down the nerve actually caused muscular 
