| 
| 
| 
1897 | LIFE HISTORY OF LILIUM PHILADELPHICUM 427 
pollen grain ( figs. 8, 9), a condition not uncommon in mono- 
cotyledons. More than a hundred such cases were noted in Z. 
tigrinum, and about thirty in L. auratum, but they are rare in ZL. 
Philadelphicum. In two cases in L. auratum a further division of 
the generative nucleus was observed, resulting in three male cells 
(fig. 10). 
There is abundant negative testimony for the usual state- 
ment that the tube nucleus never divides, but hundreds of pollen 
grains of L. tigrinum and L. auratum presented such division. 
In fig. 12 there are four nuclei, all of which have the character- 
istics of tube nuclei. One pollen grain was found with eight 
nuclei, six of which were vegetative and two generative (fig. 75). 
Numerous cases like figs. 3, 13 and rg prove that this division 
is of the direct or amitotic type. 
It may be noted in this connection that the cells of the 
tapetum often contain two, three, or even four nuclei which 
have been produced by the direct process. No evidence of 
mitosis was observed in the cells of the tapetum or in connection 
with the tube nucleus, but it is possible that it occurs in both 
cases, and that these nuclei may divide by either process. The 
significance of amitosis seems to be little understood in either 
animals or plants. The frequency of the phenomenon in patho- 
logical tissues has led to the theory that it is due to degenera- 
tion. On the other hand, such cases as the internodes of the 
_ Characeae suggest that it aids metabolism by increasing the 
nuclear surface. Its occurrence in gland cells connects it with 
extreme cell activity. If all cases of amitosis are to have the 
Same explanation it must be much more inclusive than any of 
these suggested. In Lilium a single tube nucleus seems to 
Suffice in the vast majority of cases. If amitosis is a degenerate 
Condition from mitosis, the division of the tube nucleus might 
have a phylogenetic significance. 
The pollen grains of L. figrinum often showed another varia- 
. _ tion which seems to be quite important. In fg. 17, in addition 
_ to the tube and generative nuclei, there is shown a small cell cut 
_ off sien the end d of the spore. — A similar condition is seen in 
* 
