RRinsH MARSH UIU'IIIDS 24/ 



constitution, it is inevitable that it sliould affect more tlian one 

 external character. This fact has been disguised, as Prof. Punnett 

 has privately suggested, by the fact that, in the beginning of the 

 genetic studies, some one conspicuous character was isolated for the 

 purposes of report and discussion. A more exact study would show 

 many allied external changes. But even if only a very few external 

 characters are different, and the form breeds true, we get what, if 

 classiHcation has any real value, is a distinct species. 



Where forms freely cross, as in the case of the Marsh and Spotted 

 Orchids, it may be argued that alongside the pure parents there will 

 arise a confused multitude of hybrids, with so many cross-variations 

 that no attempt to classify them is possible, and no true species will 

 emerge. We do not believe that the facts, as we interpret them, 

 support this view. In any large assembly of these plants we may 

 note, first, the occurrence of individuals, either solitary or in very 

 sparse numbers, which are certainly hybrids, but show no sign of 

 originating new strains, and, secondly, of hybrids in fairly large 

 numbers and usually of fairly uniform type, which may or may not 

 be fertile, and, in fact, established species. It may be said with some 

 confidence of most of these groups that they have been known and 

 observed in many localities ever since botany as a science has been in 

 existence, and probably they have changed but little either in type or 

 in comparative numbers. It also seems generally observed that 

 where there are hybrids present in good numbers, nevertheless one or 

 both parents are present in far greater numbers. That is, the 

 population is fairly stable, and the freest crossing never seems to 

 result in the swamping of the parent forms. That hybridization 

 does not necessarily involve a chaotic confusion of characters is made 

 quite plain in a paper by Hardy in Science, July, 1808, to which 

 Prof. Punnett has kindly called our attention, where it is shoAvn that, 

 iu a mixed population, ''there is not the slightest foundation for the 

 idea that a dominant character should show a tendency to spread ov( v 

 a whole population, or that a recessive should tend to die out." 

 A stable condition of balance is soon reached, and once reached is not 

 seriously disturbed, apart from special external conditions. See also 

 Mimicry in Buiterflies, by E. C. Punnett, 1915, pp. 154-156. Our 

 readino- of the evidence would be that we have a large number of 

 assemblages of Marsh Orchids in which some may be non-fertile 

 hybrids, resembling each other because their parents resemble each 

 other. Others may be groups of plants which are pure strains derived 

 from second generation hybrids. Others, again, may be groups Avhich 

 have originated in a mutation. In this last case, unless on the rare 

 condition of more than one identical mutation at the same time and 

 place, and a crossing of these, the new strain would emerge from the 

 second hybrid generation of a cross between the mutation and the 

 original or normal form. Thus we may have several pure strains 

 gro\ving together in various habitats and freely hybridizing. This 

 will result in a great confusion of individuals, capable nevertheless of 

 being reduced to some sort of order by careful study and comparison 

 of assemblages of plants growing in many different localities. 



