A STUDY OF RECIPROCAL CROSSES. 93 



Pistils of crookneck pollinated with golden custard, represent- 

 ing all possible stages, were picked ; beginning with those of 24 

 hours' duration and passing up to those which had been pollin- 

 ated 7 and 8 days. Pistils generally remain fresh 3 to 4 days, 

 and in some cases they did not show decay till the fifth day after 

 pollination. Of self pollinated and cross pollinated flowers there 

 seemed to be no difference in duration of flowers. In many cases 

 pistils were left on the plant till the stigma and style had decayed 

 and fallen off ; in such cases only a portion of tissue, that imme- 

 diately surrounding the ovules, could be preserved, but this was 

 actually necessary in order to give pollen tubes — if they did 

 germinate and enter the ovarian region — a chance to act, and 

 furthermore to give the investigator an opportunity to study the 

 final action of such tubes. 



Of course a study of the germination of pollen tubes necessi- 

 tated a large amount of histological work, and in order to get 

 material in condition for this study the following method was 

 followed : Pistils were picked, trimmed so as to avoid all 

 unnecessary tissue, killed, and fixed in a one per cent solution of 

 chromo-acetic acid for 30 hours, washed an equal length of time 

 in tap water, hardened and dehydrated in alcohol of 35, 50, 70, 

 95 per cent solution, and finally in absolute alcohol. The alcohol 

 was replaced by xylol, and the pistils were then imbedded in 

 paraffin. Sections were cut at thicknesses varying from three 

 and one-third to twenty micro-millimeters ; twenty micro-milli- 

 meters proved the most satisfactory. The three-color-stain — 

 Flemming's safranin, gentian-violet, and orange G. was used. 

 The tissues were mounted in balsam and finally studied with the 

 microscope. 



The object of making and examining these sections of the pis- 

 tils and ovaries was to ascertain if possible, some reason why 

 fecundation failed to occur, the chief aim being to trace the pol- 

 len tubes, if formed, from stigma to ovules. The tracing of pol- 

 len tubes is a comparatively easy thing to do, provided the tissue 

 is properly handled and successfully stained. The triple stain 

 previously mentioned proved especially good for this work. In 

 this case the cytoplasm of embryo-sac and pollen tube stains from 

 gray to orange, the nucleus violet, and nucleolus red. Besides 

 being different in color, the pollen tubes are distinguished by the 

 peculiar granular structure of their protoplasm. 



