1916] Brandt: Notes on Trillium 51 



manent effect than the checking of a bud less advanced. This outer- 

 most bud, in the relatively few cases when it thus actually develops, 

 tends to produce a smaller flower and seems less likely to produce a 

 fruit than the next. This statement is based upon field observations. 

 Because of the great increase in size of the buds for the next season 

 it becomes very easy from this time on to distinguish those whose 

 development has been arrested. 



Somatic divisions are taking place in all parts of the young buds, 

 but more especially in the anthers. From the middle of May to the 

 middle of June the archesporial cells of the anthers are cut off (pi. 7, 

 fig. 8). In an anther of T. sessile var. giganteum there are usually 

 four rows of archesporial cells. These closely resemble the adjacent 

 cells, but may be recognized by their larger size, radial elongation, and 

 position immediately under the epidermis. They soon begin to divide 

 to form sporogenous and parietal tissue. As development progresses 

 the differences between sporogenous and parietal cells become more 

 and more noticeable. All the parietal cells which are arranged in 

 layers are greatly flattened as a result of periclinal division, their 

 cytoplasmic content is not very abundant and their nuclei are rela- 

 tively small but rich in chromatin. The sporogenous cells, on the 

 other hand, are not arranged in layers and have no definite form. 

 They are non-vacuolated and their cytoplasm shows a reaction to stains 

 different from that of the cytoplasm in other cells of the anther. Their 

 nuclei are large, but rather less densely filled with chromatin than 

 those of the parietal cells. There is therefore a great difference in 

 appearance between sporogenous and parietal cells, but the tapetum 

 is not yet cut off. Late in July the ovules begin to make their ap- 

 pearance as slight protuberances of the placentae, but their arche- 

 sporial cells do not appear until later in the year. 



h. Trillium ovatum 

 Practically the same series of phenomena reported above for T. 

 sessile var. giganteum- occurs during this active period below ground in 

 T. ovatum, but a few exceptions should be noted. Mention has been 

 made of the fact that fewer buds are formed in T. ovatum, the number 

 being limited to two in average cases. It might here be noted that 

 the peduncle becomes noticeable before the flower-bud has attained 

 one-fifth of its mature length. The most noticeable difference is in 

 the rate of development of the floral organs. In some specimens 

 young pollen grains are present before the last of July. This very 



