144 FERTILIZATION OF SPRING FLOWERS ON THE YORKSHIRE COAST. 



but in the Scarborough plants there are four very obvious nectaries at 

 the base of the flower. The stamens shed their pollen upwards and 

 inwards at a distance of 1 mm. from the stigma, so that in the 

 homogamous flowers an insect seeking honey, unless it be very 

 small, will simultaneously touch with one side of its body a stamen 

 and with the other the stigma. Small insects like Sepsis leave the 

 flower abundantly covered with pollen. The plant is restricted to 

 the most windy, exposed headlands on the coast ; on these, how- 

 ever, it abounds. 



Capsella Bursa-pastoris. — It has been suggested that the gyno- 

 dioecism and gynomonoecism of this plant is brought about by cold.* 

 My Scarborough observations very strongly support this contention. 

 The plant soon came into flower after the severe frost of January 

 and February, '95, but every flower on hundreds of plants which were 

 examined bore only rudimentary stamens until April 9th, upon and 

 after which ? flowers began to appear all over the district.! In 

 '96, after a mild winter, the earliest flowers of all the early-flowering 

 inflorescences were observed to be ? , but ^ flowers began to appear 

 towards the end of March. In consequeuce the plant did not lose 

 much by the absence of visitors, for with or without them fertili- 

 zation was impossible. Sex- separation such as this is in no way an 

 adaptation leading to cross-fertilization, as is that in many more 

 highly-specialized flowers. 



* Willis, " On Gynodioecism (third paper), with a preliminary note upon 

 the origin of this and similar phenomena," Proc. Cambridge Phil. Soc. viii. 

 (1893), p. 130. It may be noted that the plants spoken of in this paper " had 

 just begun to flower." 



t The same state of things was noted in Germany in this year. See 

 Warnstorf, "Bliithenbiologische Beobachtungen aus der Euppiner Flora in 

 Jahre 1895," Abhandl. d. hot. Vereins Brandenburg, xxxviii. 1896, p. 15. 



