Charles J. Chamberlain 9 



as a centrosome was found in our material, which furnished a series from the initial 

 cell up to stages in which more than thirty cells appear in a transverse section 

 of the antheridium. Unfortunately, the material showing the last two or three divi- 

 sions preceding the formation of the spermatozoid mother-cells was not satisfactory, 

 and, consequently, no positive statement can be made in regard to blepharoplasts, 

 although we should assume them to be present during the last one or two mitoses. 



In the germinating spore a differentiated area, already described as a centrosphere 

 by Farmer (9), Strasburger (33), and by Davis (7), is often found at the center of the 

 aster. The origin and behavior of this structure, which we regard as a genuine cen- 

 trosphere, are rather puzzling. While we assumed that it must appear earlier than 

 the rays, and that the rays were developed from it, the failure to identify the structure 

 before the appearance of the rays, and its frequent absence when it might be expected 

 to be present, led to a careful study of the subject. The conclusion was reached that 

 the centrosphere gives rise to the rays, but that the rays may also contribute materi- 

 ally to the substance of the centrosphere. 



Although we have not been able to make any satisfactory study of living material, 

 we believe that appearances warrant the theory that there is a streaming movement in 

 the rays. Such a theory is not entirely new to zoologists. If the theory be true, 

 when the streaming is toward the nucleus the centrosphere would increase in size, 

 while a continued streaming toward the periphery would cause the centrosphere to 

 disappear. In regard to the origin of the rays, nothing more definite was determined. 

 Finely granular areas, showing a tendency to stain with gentian violet, were sometimes 

 seen in earlier stages, but the actual formation of rays or centrospheres from these 

 areas could only be surmised. These areas do not seem to differ essentially from those 

 which we (4) have already observed accompanying the male nuclei of Pinus Laricio. 

 In some of Miss Ferguson's (13) figures of the same species and of Pimis rigida the 

 areas approach the form of definite centrospheres. The aster appears so suddenly that 

 its mode of development is largely conjectural. In a fully developed aster, there is 

 usually an increase in the diameter of the ray at the centrosphere (Plate XXVI, Figs. 13 

 and 16), and occasionally a slight enlargement at the Hautschicht An enlargement of 

 the ends of the rays, as shown in Fig. 13, is just what should be expected if there is a 

 streaming of material. The variability in the size of the rays and their irregularly 

 granular character also favor the theory that they are lines of streaming material. The 

 tendency of small nucleoli or microsomes to collect on the rays, as pointed out by 

 Schaffner (27) in his study of Lilium, and as is familiar to all who have seen mitoses in 

 the embryo-sac of Lilium and similar forms, is another argument in favor of this theory. 



The asters arise at opposite poles of the nucleus, but not simultaneously. Serial 

 sections of a large number of nuclei were examined before this conclusion was reached. 

 We can hardly understand Davis's (7) statement that in his studies he "has never 

 found a nucleus with a clearly defined solitary aster beside it. This is a very impor- 

 tant point and the search was persistent." In our own preparations of the second and 



335 



