THE ORIGIN OF SPECIES 



slowly than normal. As a result, if species A is pollinated both by A and 

 by B pollen, practically all of the fertilizations will be conspecific {A X A) 

 rather than heterospecific (A X B), because A pollen on A styles grow at 

 a normal rate, while B pollen on A styles grow at a reduced rate. While 

 this phenomenon is well known in interspecific crosses, it is not unknown 

 in intraspecific crosses. Thus, sugary and starchy maize differ only by a 

 single gene, yet the growth rate of sugary pollen tubes on starchy silks is 

 slower than that of starchy pollen tubes. The two types of pollen tubes 

 have equal growth rates on sugary silks. 



The bursting of pollen tubes is particularly likely to occur in crosses 

 in which the male parent has a chromosome number greater than that of 

 the female parent. Wherever the chromosome numbers of the species are 

 equal, the ratio of the chromosome number in the style to that in the 

 pollen tube will be 2 : 1 ( diploid to haploid ) . But when the ratio comes 

 closer to 1:1, then bursting of pollen tubes commonly results. This is 

 particularly likely to occur in crosses between polyploid species (see 

 Chapter 19) and their diploid ancestral species. For example, it appears 

 that the commercial tobacco, Nicotiana tahacum, with a chromosome 

 number of 48, is derived from N. sylvestris and IV. tomentosa, both of 

 which have 24 chromosomes. The cross of either of the latter species to 

 tahacum can be made easily, provided that tahacum is used as the female 

 parent. This will give a ratio of stylar to pollen tube chromosomes of 4 : 1. 

 But if the same cross is attempted using tahacum as the pollen parent, the 

 ratio is 1 : 1, and the cross usually fails because of bursting of pollen tubes. 

 It has been shown that, in some cases, the bursting of pollen tubes results 

 from a greater osmotic pressure in the pollen tube than in the style. While 

 this is correlated with chromosome number, it is clear that the higher 

 osmotic pressure of the pollen tubes is a physiological effect of the geno- 

 type rather than a direct effect of the number of chromosomes, for chro- 

 mosomes never exist in sufficient numbers to have important effects on 

 osmotic pressure. Nothing strictly comparable is known in animals. How- 

 ever Serebrovsky has shown that pH and osmotic pressures of vaginal 

 secretions of various domestic mammals have small but consistent dif- 

 ferences. Interspecific inseminations generally result in death of the sperm, 

 and this may be a result of osmotic incompatibility. 



Once a hybrid zygote is formed, there is no assurance that it will reach 

 maturity, for death may occur at any stage during development. Thus in 

 intcrfamilial crosses of sea urchins, the paternal chromosomes are ex- 

 truded, and the resulting haploid larva dies at an early stage. In tlie cross 

 between Datura stramonium and D. metel (Jimson weeds), development 

 proceeds to the eight-cell stage, then stops. In many plant hybrids, lethal- 

 ity appears to result from an inadequate nutritive relationship between 

 the endosperm and the embryo, for embryos which would otherwise die 

 may be brought to maturity if they are dissected out of the seed and raised 

 on an artificial nutritive medium. The adult plants so obtained may be 

 fully as vigorous as the parent species. 



Little progress has been made in analyzing the genetic basis of inter- 

 specific sterility, and for this reason, the two cases discussed below have 



278 



