A MORPHOLOGICAL STUDY OF ULMUS AMERICANA. 
CONTRIBUTIONS FROM THE HULL BOTANICAL LABORATORY. 
LXXVIII. 
CHARLES H. SHATTUCK. 
(WITH PLATES VII-IX) 
A stupy of this species was suggested by the interesting results of 
recent investigations in this region of the Archichlamydeae; notably 
those of KarsTEN (12), Miss BENSON (1), ZINGER (22), and NAWAS- 
CHIN (16). 
METHODS. 
Collections were made from February 13 to May 1, 1903, and 
tepeated during the same period for 1904. During the first year 
collections were made on.alternate days throughout the more rapid 
period of growth, and at intervals of ten days at other times. During 
the second year the same plan was followed except that collections 
were made every day during the period of fertilization and embryo 
growth, | 
The ovules are covered with a dense growth of hair which pre- 
vents sinking in the killing fluid; but after immersion in 95 per cent. 
| they sink at once. A 2 per cent. solution of chromo- 
acetic acid was found to give the best results as a killing agent for all 
but the oldest stages;. these requiring a somewhat stronger solution. 
The material was imbedded in paraffin and the sections were cut 
from 2 to to in thickness. A preparation of Le Page’s glue and 
Byeerin was used for fixing sections to the slide (glue 40 parts, 
Hee 10 parts, glycerin 50 parts). The albumen and several other 
The ne oe oa ted, but all failed to fix the sections to the ae: 
shige ‘3 — 1s as perfectly transparent on the slide as Mayer's 
sive, will k hag cag easily prepared, is a much stronger adhe- 
Th eep indefinitely, and is not so easily coagulated by heat. 
2... = sactory combination for staining the ovules was 
to the abo patranin and gentian violet. The addition of orange G 
1905] ve brought out the pollen tubes best, as they hold the gen- 
209 
