1916] Brandt: Notes on Trillium 53 



early part of September. The anthers are at this time abnost fully 

 grown and nearly fill the interior of the flower-bud. The pollen-sacs, 

 which are still nearly colorless, are also very slender, owing to the 

 fact that the pollen mother-cells are not rounded off. They are very 

 angular in outline and, when crushed out of a living anther, cohere 

 firmly in a mass. The cytoplasm is dense and the nuclei are verj^ 

 large. From one to three nucleoli are present and the open reticulum 

 is composed of thinly scattered chromatin granules (pi. 8, fig. 1). 

 The tapetum is definitely cut off, but its cells have not as yet become 

 so greatly enlarged as they appear in a mature anther, nor have their 

 nuclei divided. 



The pollen mother-cells continue in the resting stage for four 

 weeks or longer before rounding off and preparing to divide. At 

 about the middle of September or early in October, when the rainy 

 season begins and the weather becomes cooler and the soil contains 

 some moisture, the underground organs of Trillium sessile var. 

 giganteum begin to show increased activity. This increase is manifest 

 first in a lengthening of the protective sheaths of the crown, and then 

 in an increase in size in all the parts of the stem buds for the next 

 active season. There is a very noticeable change in the appearance 

 of the anthers. Their length does not increase very appreciably, but 

 the pollen-sacs begin to increase rapidly in diameter and their color 

 begins to change from nearly white to pale yellow — the color of pollen- 

 sacs in which the reduction divisions are taking place — and finally, 

 as the pollen attains maturity the color becomes deep yellow. When 

 examined under the microscope the pollen mother-cells are found to 

 be rapidly increasing in size and their nuclei show a more than pro- 

 portionate increase (pi. 8, fig. 1). In some locules pollen mother-cells 

 are in synapsis, while in others, even in the same anther, the chromatin 

 is still distributed in the form of rather large granules throughout the 

 nuclear cavity. These "resting" nuclei are fully as large in diameter 

 as are those in synapsis (see pi. 8, figs. 1 and 2). I have not found 

 such an increase in size of the nuclear cavity during synapsis in 

 Trillium as Lawson (1911) reports in Smilacina. 



The statement has already been made that in the same flower-bud, 

 or even in the same anther, in T. sessile var. giganteum the pollen 

 mother-cells in one locule may be in synapsis, while in another the 

 chromatin still forms an open reticulum. It may be well here to state 

 that the period during which the meiotic divisions occur extends over 

 two months — October to November — a condition in striking contrast 



