A CONTRIBUTION TO THE PROBLEM OF 
HOMOTYPOSIS 
DATA FROM THE LEGUME CERCIS CANADENSIS 
By J. ARTHUR HARRIS, Ph.D., Carnegie Institution of Washington, U.S.A. 
I. Introductory Remarks. 
In 1905, I began the collection of series of data for a comprehensive study 
of the problem of intra-individual or "homotypic"* correlation for fertility and 
fecundity characters in plants. Such characters were chosen not merely because 
of particular interest in that subject in general but because of the conviction 
that the intra-class and inter-class f correlation methods applied to the problem 
of fertility and fecundity might yield results of considerable physiological interest. 
Since that time, circumstances have prevented my carrying out the work along 
the lines originally laid down. The constants given here are drawn from notes of 
the work then done. 
II. Materials. 
The data presented are exclusively those for pod length I, number of ovules 
formed o, number of seeds developing s, and number of ovules failing to develop 
/, in the small arborescent legume, Cercis Canadensis. Three collections are 
involved: one from Meramec Highlands, near St Louis, Mo., made in the autumn 
of 1905; one from the vicinity of Lawrence, Kansas, taken in 1905; one from 
the neighbourhood of Sharpsburg, Athens County, Ohio, gathered in 1908. These 
materials have already been considered in dealing with other problems quite 
distinct from the present one ; the reader seeking more comprehensive information 
with regard to the materials used should consult the papers cited below J. 
For the Meramec Highlands collections two sets of symmetrical homotypic 
correlation tables were prepared. 
* K. Pearson, and others: Phil. Trans. B, Vol. cxcvii. pp. 280-288, 1901. 
t Biomeirika, Vol. ix. pp. 440-472, 1913. 
J Boi. Gaz. Vol. l. pp. 117-127, 1910; Bull. Torr. Bot. Club, Vol. xli. pp. 243-250, 1914 
