FRUIT FORMATION 481 



especially sensitive to sugar solutions, growing toward them readily. 

 They tend to grow away from dry air and "show a preference for spaces 

 saturated with aqueous vapour to such as are less humid. "^^ Investi- 

 gations of the mode of growth of the pollen tube in Houstonia led to the 

 conclusion that the tissues of the style influence its direction only in a 

 passive manner but that "a chemotactic stimulus originating in the 

 egg-apparatus, or the egg itself, is the chief directive influence, "^^ 

 Dorsey, however, has found tubes growing in plum styles with aborted 

 ovules; therefore it is possible that growth often depends less on a 

 normal egg-apparatus than the work with Houstonia would indicate. 

 Dorsey found also that in the apple the pollen tube may grow beyond 

 the ovule and down into the stem. Kerner and Oliver^^ state that 

 ovules ready for fertilization "attract not onlj^ pollen-tubes from pollen 

 of the same species, but of others far removed from it in point of affinity. 

 The delicate hyphae of several mould-fungi are similarly attracted." 



Time for Pollen Tube Growth. — Ordinarily germination of the pollen 

 grain occurs promptly after pollination, the pollen tube grows fairly 

 rapidly and fertilization occurs within a period of 1 or 2 days, though 

 the time may be expected to vary with temperature and other environ- 

 mental factors. Under favorable conditions there is an interval of from 

 9 to 120 hours between pollination and fertilization in apples, plums 

 and cherries."'*'**^ The very much slower growth of Rome pollen tubes in 

 Rome styles as compared with that of the tubes of other apple varieties 

 found by one investigator^^ is interesting and may offer an explanation of 

 some cases of self sterility. A period of from 26 to 41 hours has been re- 

 ported in the case of certain cucurbitaceous plants, ^^ 4 days in one of the 

 species of Gastrodia,^"^ one month in Betula, ^ several months in H amamelis^^^ 

 and approximately a year in certain of the oaks.-^ That there may be 

 a great variation in this respect between closely related plants is evident 

 from the behavior of the Satsuma orange in which about 30 hours have 

 been found to elapse • between pollination and fertilization,^^ while a 

 corresponding period of 4 weeks has been reported in Citrus trifoliata.^°° 



Fertilization. — In Fig. 13 of Plate II are shown the vegetative 

 nucleus and the generative cell of the mature pollen grain. During the 

 growth of the pollen tube the nucleus of the generative cell divides, giving 

 rise to two male gametes, each consisting of a nucleus and a small portion 

 of stainable material. The pollen tube, after entering the micropyle, 

 penetrates the intervening tissue of the nucellus and then enters the 

 embryo sac. The following account of fertilization is adapted from 

 Mottier's^^ description of the process: The end of the tube may enter 

 the sac at one side of the synergids, in which case only one of these cells 

 is at once disorganized, the other retaining its normal structure for some 

 time. This condition is illustrated in Fig. 5, Plate I. Often it enters 

 between the two synergids, in which case both cells disintegrate almost 



