BRITISH JrARSII ORCHIDS 24:0 



experimental crosses o£ various kinds, and the raising of the progeny 

 of those crosses, if fertile, through at k^ast three generations. The 

 ditiicidty here is hoih the uncertainty of germination at tlie outset 

 and tlie slow growth to the flowering stage. This has been set down 

 as seven years ; but that period would probably be much shorter 

 under favourable conditions. In any case it would seem that the 

 experimental work could best be carried on at some plant-breeding 

 institution, where continuous work through many years might be 

 assured. It would not be enough, by a cross, to get some plant very 

 near to O. latifulia. It would be necessary to find out whether 

 this form were fertile ; if it were, there is no reason why it might 

 not establish itself in numbers and perpetuate itself as a separate 

 species. 



In the absence of guiding experiments, we are bound to discuss 

 the origin and connections of these ty])es with some ideas in our 

 minds as to the manner in which one form has arisen out of another. 

 Pi-esumably we all believe either in evolution (in the strict sense) or 

 in epigenesis, and for the purposes of this study the}^ mean the same 

 thing. We are bound to have some theory of the origin of t^^pes so 

 fully segregated as say 0. ustitlata and O. morio, and such types as 

 the Marsh and Spotted Orchids, in a state of " polvmorphic mixture." 

 For this we are obliged to fall back upon Mendelian study and to 

 seek what assistance we can get from the laws of mutation and 

 segregation so far as they have been elucidated. Perhaps not much 

 help is here to be expected ; but we may at least be able to deter- 

 mine what is possible, if not what is probable or necessar}^. 



In these matters the opinions of botanists seem to be very much 

 in chaos. One will deny that mutation is a vera causa of new 

 species, and another that crossing may so result ; and between the 

 two we are left with no theory at all. One will say that mutations 

 only affect single characters and species are built up on numerous 

 characters ; or that if you get hybridization at all freely, the result 

 is a jumble of polymorphic forms, out of which no species can be 

 distinguished, and in the midst of wdiich nothing is stable. This 

 would seem to lead us straight back to the old position that if two 

 forms crossed they must be reckoned to belong to the same species. 

 Here we might do well to quote a sentence or two from Bateson's 

 Presidential Address to the British Association at Melbourne in 

 1014 : — "Who could have foreseen that the Apple and the Pear — so 

 like each other that their botanical differences are evasive—could not 

 be crossed together, though s]:>ecies of Anfirrhimim so totally unlike 

 as majus and molle can be hybridized without a sign of impaired 

 fertility ? " — and then, " The only definable unit in classification is 

 the homozj^gous form which breeds true " (p. 13). We know as yet 

 \tery little, if anything, about the conditions of compatibility wdiere 

 crossing occurs : in the meantime it seems to us most important to 

 keep in mind the possibility that in cases of polymo^-phic mixture we 

 may have true-breeding races which also freely cross with other 

 species. The problem is to distinguish the pure from the hetero- 

 zj'gous forms, Avhere all are growing in close association. 



