404 FERTILIZATION AND FORMATION OF FRUIT IN PHANEROGAMS. 



pollen-tubes, sometimes found, though in some cases they penetrate the tissues of 

 the stigma, do not fertilize the ovules. It is another question whether or no this 

 "foreign" pollen is entirely without effect, whether it does not possibly influence the 

 stigmatic tissue so that less foreign pollen, arriving later on the same stigma and 

 developing pollen-tubes, is affected. But this subject can only be dealt with later 

 on; here it suffices to state that all pollen falling on a stigma is not necessarily 

 suitable, and that the stigma has, in a manner of speaking, to make a selection. 



It is very difficult to say what conditions come into play in this choice of pollen. 

 Experimental inquiry into this matter has not been wanting, but its results tell us 

 little as to the fundamental processes at work. By it we ascertain little moi'e than 

 whether this or that artificial pollination leads to a production of seeds or not. 

 Thus in one case no seed will be formed, in another a few, and in a third case an 

 abundant crop. The sources of error in this class of experiment are considei'able, 

 nor do the results b}^ any means always harmonize. Thus, in experiments of my 

 own as to the fertility of certain Catchflies when pollinated from allied forms, no 

 result would be obtained in one year, whilst in the following year their repetition led 

 to the production of a certain number of seeds. Other observers have had the same 

 experience; and it would seem that whatever care be exercised, absolute reliance 

 cannot be placed on the result — especially where it is a negative one. Caution must 

 be used, therefore, in generalizing from such experiments, especially in cases where 

 their number is limited. In the main, the general results are very instructive, and 

 must not remain unnoticed here in so far as they relate to the connection between 

 fertilization and the origin of new species. 



These results may be shortly summarized as follows. When the pollen of one 

 species is placed on the stigma of another species, pollen-tubes capable of fertilizing 

 the ovules are developed only when the two species belong to the same genus or to 

 the same natural family of plants. Families and genera are conceptions devised by 

 Botanists, and although their limitations are to some extent arbitrary or dependent 

 on the personal equation of individual observers, in the main there is little differ- 

 ence of opinion as to these limitations in the case, at any rate, of families. How far 

 new discoveries may lead to a revision of their present limits must remain un- 

 decided, but, of families as at present laid down, we may say that crossings of pollen 

 between species of two different families (orders) is without result, whilst between 

 species of two different genera very rarely is seed produced. 



The crossing of species of the same genus results, in most cases, in fertilization, 

 and eventually in the production of hybrids. It is certainly remarkable, in this 

 connection, that external similarity between the two species crossed has little bear- 

 ing on the result or absence of result. One of the commonest of naturally-produced 

 hybrids is one which owes its origin to the union of Primula glutinosa with 

 Primula minima, two species very dissimilar in the form of their foliage and 

 flowers. On the other hand, hybrids of the very similar Cowslip and Bardfield 

 Oxlip (Primula officinalis \yeris] and Primula elatior) are but rarely met with in 

 nature, whilst artificial pollinations between them only occasionally lead to any result. 



