1913] MARTIN—POLLEN OF TRIFOLIUM PRATENSE 125 
and that the endosperm cell had made one division; therefore, 
fertilization must take place about 50 hours after cross-pollination. 
An examination of those flowers self-pollinated at the same 
time the others were crossed showed good germination on the 
stigma. The number of pollen grains germinating on their own 
stigma ranged from 3 to 25 in the 30 flowers. The tubes produced 
were all short, none exceeding 4 mm. Out of 20 self-pollinated 
flowers run go hours, one tube was found with a length of 7.25 
mm.; the other tubes varied in length from a fraction of a milli- 
meter up to 5 mm. Counting the average length of the style and 
stigma 11.5 mm., one is able to compare the rates of growth of 
- pollen tubes in the two cases. The tubes in case of self-pollination 
look as vigorous as those in cross-pollination. Some abnormal 
behavior was observed. In one case the tubes were found wound 
about each other in the upper part of the stylar canal. In a few 
cases one of the longer tubes had turned back upon itself. 
The question is now raised in case of self-pollination as to 
whether or not the tube can reach the ovary and effect fertilization. 
Field work on self-pollination shows that it rarely does, if ever. 
Summary 
The pollen of Trifolium pratense is physiologically different 
from that of T. hybridum and T. repens in respect to behavior 
toward sugar solutions. 
The only function of the sugar solution in the case of the pollen 
of T. hybridum is the controlling of water supply. 
The germination of the pollen of 7. pratense is delicately adjusted 
to water absorption. 
The results of the investigation show that the stigma produces 
no secretions which influence pollen tubes. 
The nature of the pollen demands no other function of the 
stigma in its germination than the control of the water supply. 
The pollen in self-pollination germinates readily on the stigma, 
but the tubes traverse the style much more slowly than in cross- 
pollination. 
In conclusion I wish to express my thanks to Dr. WILLIAM 
Crocker and Dr. Soputa EcKERSON of the University of Chicago, 
