384 J. M. D. OLMSTED 
The conduct of the phagocytic leucocytes in degenerating 
taste buds and their presence in normal buds accompanied by a 
surrounding phagocytized area, is clear evidence that this is the 
method of ridding the normal bud of worn-out sense cells. <A 
bud was found in a normal barbel in which the leucocyte seemed 
to be attacking one cell only. For some distance above and 
below the included cell this one particular taste cell was entirely 
devoid of cytoplasm, a few coarse granules, such as are usually 
found within a wandering pigment cell, being left in the more 
distal portion. The phagocyte was distinctly amoeboid in shape, 
as if it has been fixed in an active state. In several normal buds 
where the wandering cells were evidently in the act of phago- 
cytizing the sense cells, the picture was exactly like that in one 
of the barbels whose nerve had been cut thirteen days. In the 
latter specimen, however, 1t was the beginning of the final destruc- 
tive process, for other buds were seen in a much more advanced 
stage of degeneration. Figure 3 illustrates the most advanced 
condition in a normal bud I have ever happened upon. Gener- 
ally the process goes no further than that indicated in figure 2. 
If phagocytosis to such a degree as shown in figure 3 was observed 
by Baginsky in normal dogs and on the side of the tongue corre- 
sponding to the unoperated ninth nerve, it is not surprising that 
he decided there was no change from the normal condition after 
cutting the nerve. But why the buds should have remained 
even in this condition for eighty-seven days in his experiment is 
inexplicable, that is, if the operation had been correctly performed. 
In normal taste buds new sense cells most probably arise from 
the germinative layer of the epidermis, as they do in regenerating 
barbels, and not by a division of the already existing ones, since 
mitosis is never seen in these cells. This is a further argument 
against dedifferentiation, for if these cells cannot even divide 
to form other cells like themselves, but are destroyed when they 
are worn out, how much less would they be able to change back 
to a kind of cell which is in a more primitive condition and which 
possesses the possibility of differentiating back again into that 
same more specialized sense cell. I believe more careful and 
extended experiments on the dog will show the same results as 
