MENDEL'S "LAW APPLIED TO ORCHID HYBRIDS. 



689 



above are abnormal and exceptional, owing to the peculiar kinds of 

 elements made use of, and therefore, for the present, they may be 

 disregarded altogether. 



Yet, notwithstanding this, the results of Mendel's experiments in the 

 second generation are highly important, and suggest a formula which I 

 believe may be greatly extended in its application. 



Unfortunately, Mendel's formula and Mendel's theory in explanation 

 of that formula have hitherto been regarded together and put forward as 

 the same thing, under the name of Mendel's law. It has already been 

 shown in a large number of cases that Mendel's formula is an undoubted 

 reality as a working formula for the separation of characters in self- 

 fertilised crosses. 



In the following observations in regard to Orchid hybrids I propose to 

 show that this working formula of Mendel may, with certain modifica- 

 tions, be extended to the union of specific characters in intermediate 

 primary hybrids. 



At the same time I have reason to believe that the application of 

 Mendel's formula is strictly limited to hybrids and crosses of a certain 

 ancestry. I hope to show, further, that Mendel's theory, in explanation of 

 his formula, cannot generally be maintained with Orchid hybrids ; and, 

 finally, that Mendel's formula is not a general law, but simply a particular 

 formula for a particular group of hybrids and crosses. 



First, in regard to the extension of Mendel's formula to the union 

 of specific characters in primary hybrids. 



In order to clear the way for the details of my observations in Orchid 

 hybrids, I will offer a few remarks on the nature of those observations, 

 which may possibly serve to anticipate and meet several objections similar 

 to those that have been already made to the experiments of Mendel and 

 others. 



The Advantages of Okchid Hybeids for Observations in 



Heredity. 



My own experiments and observations in hybridisation have, for the 

 most part, been deliberately confined to hybrids between distinct species 

 in the natural order Orchidece ; these seem to me to offer distinct advan- 

 tages to the student of heredity, for the following reasons : — 



1. The natural order OrchidecE contains a large number of genera and 

 species, with a wide range of variation. 



2. Many of the species have attained to a high degree of specialisation 

 among flowering plants. 



8. Orchid hybrids, between distinct species and even genera, are now 

 very numerous in gardens, and, as a rule, are very fertile as compared 

 with some plant hybrids. 



4. Orchid hybrids being valuable commercially, their pedigrees are, as 

 a rule, carefully kept ; and the literature devoted to them is very exten- 

 sive, containing not only written records, but also valuable reproductions 

 of paintings, photographs, and drawings, all of which are, and will be, 

 invaluable for future reference. 



5. Orchid hybridisation has progressed so rapidly of late that we have 

 already complex hybrids combining in their pedigree no less than five 



F 



