248 Monolepsis in Orchids [ch. 



discussion of unexpected genetic phenomena this possibility 

 must be taken into account"^. 



There is a further difficulty to be considered in this 

 connection. The fact that no seed is set when fertilisation 

 is excluded is in itself no proof that the inheritance is not 

 of a parthenogenetic type. A small but very definite and 

 well-ascertained group of cases occurs among Orchids in 

 which, though no seed is set unless the flowers are pollinated, 

 yet the offspring exhibit no trace of the paternal characters. 

 The facts are well known to orchid-growers, but they were 

 first made known to students of genetics by Hurst (152, 

 p. 104). The most notorious example is that of Zygopetahun 

 Mackayi, which when fertilised with pollen of various other 

 orchids has given nothing but Zygopetaltun Mackayi. Four 

 species of Odontoglossiim, a Lycasta, and an Oncidium, used 

 as father all gave this curious result. Hurst quotes other 

 records showing that several other orchids may behave in 

 the same way as Z. Mackayi, giving purely maternal plants 

 when fertilised with pollen of other types. In some of these 

 cases it has also been shown that self-fertilisation of the 

 supposed cross-bred again gives nothing but the same type. 



Pending cytological investigation it is not possible to 

 form a satisfactory rationale of these cases, but we cannot 

 avoid the conclusion that fertilisation has acted as a stimulus 

 to development without any admixture of paternal character. 

 The result is, as Hurst has pointed out, exactly as if 

 parthenogenesis had in reality occurred, although pollination 

 had taken place and is indeed necessary to effect production 

 of seed. I have proposed to distinguish this phenomenon 

 as Mo7iolepsis, in which characters are taken from one 

 parent only, in contrast to Ampkilepsis, in which they are 

 brought in from both sides. 



In the earlier staofes of these discussions we were dis- 

 posed to attach much importance to an observation of 

 Millardet (198), to the effect that in certain crosses among 

 strawberries the supposed hybrid was of purely paternal 

 type, showing no maternal features, and breeding true. 

 The term ''false hybrids" was suggested by him as descrip- 

 tive of such plants. No exactly similar example has been 



* For important summary of new evidence see Rosenberg, Svensk Bot. 

 Tidsk., 1909, HI. p. 159. 



