122 MISSOURI BOTANICAL GARDEN. 



Tables for the correlation between the number of flowers 

 per inflorescence and the number of seeds per fruit, and for 

 the number of fruits developed per inflorescence and the 

 number of seeds per fruit, are given under the numbers II 

 and III. The possibility of some uncertainty concerning the 

 number of flowers formed will be borne in mind. 



In these tables the numbers of flowers formed and fruits 

 developing are necessarily weighted with the number of 

 fruits which they actually produce. In calculating the cor- 

 relation coefl^cients, standard deviations based on these 

 weighted frequencies were calculatetl, but it seems unneces- 

 sary to publish them here. The constants for number of 

 seeds per fruit are: 



^^^^ 4.0788 ±.0172 



Standard deviation 1.2817±.0122 



Coefficient of variation 31.422 



The relationship between number of flowers per inflores- 

 cence and number of seeds per fruit, and between number of 

 fruits developing per inflorescence and the number of seeds 

 per fruit are: 



Number of flowers and number of seeds, r= .033 ±.013 

 Number of fruits and number of seeds, r= — .012 ±.013 



Possibly the first of these coefficients is significant with 

 regard to its probable error; the second is certainly not. In 

 the calculation of these probable errors, A'' was taken as the 

 actual number of fruits involved in the correlation table. If 

 we consider that the degree of trustworthiness of our work 

 is determined rather by the number of inflorescences, our 

 probable errors will be so raised that neither of these' con- 

 stants can possibly be regarded as significant. The fact 

 that one is positive and the other negative also raises some 

 doubt as to their significance. 



We conclude therefore that, so far as the data now avail- 

 able for this species enable us to determine, there is neither 

 sensible interdependence between the number of flowers 

 formed on an inflorescence and its capacity for maturing the 

 ovaries of these flowers into fruits nor sensible relationship 

 between the size of an inflorescence as measured by the num- 

 ber of flowers it produces and the number of seeds per fruit. 



