62 Cooperative Investigations oh Plauts 



If instead of collecting sced from a singlo capsiile wc collect all tlie seed from 

 all the capsiiles of a plant, \ve may still further wideii our rauge to 20 grades of 

 sibship, and it is not by aiiy iiieaus certain that these 20 grades will give as 

 close a general rclationship, as the grades from the same capsule even if the 

 plant and not the flower be the individwal. For, the more capsules taken the 

 more chance there is for variety in the crossing with other plants, i.e. there is 

 increased aihnixture of the half-sibliiig relationships. 



It is further obvious how wide must be the ränge of experinient if \ve are to 

 deterniiiie the relative influenae of these various niodes of crossing; for, every 

 heredity series to have validity ought to contain at the very least 100 plants, 

 and the majority of the cases discussed require artificial fertilisation, and in some 

 of the inctatropic cases this is of a rather complex kind. 



The series of experinieiits considered in this paper do not deal with the 

 questiun of fertilisation at all, althoiigh tentative experiments in this direction 

 have been coMHuenced this year. Shirley poppies grown in nuisscs are both seif- 

 and cross-fertilised, the latter principally liy bees altliough the flowers are not 

 hoiieyed. The relative extent of euch form of fertilisation eould probably not be 

 determined without special investigation or experinient*. To what extent the 

 cross-fertilisation is really endotropic and not exotropic, in the c:xse of plants 

 having sometimes a dozen Howers in blossem at the same time must also be the 

 subject of a later enquiry. It is clear, however, that the relative proportions 

 of hoinotropy, endotrop}' and exotropy may vary very largely from one crop to a 

 second and thus largely influence the ' parental ' and ' fraternal ' relationships. 



ütlier matters directly iufluencing the constants are, besides soil and climate, 

 differeutiation between early and late Üowers and between high and low tiowers 

 (which is partly the same thing), aml tlie etfect of special treatment of individual 

 plants. \Ve shall refer to these topics again later. 



(3) Influence of Environment on Grop-constants. It must be remembered that 

 we had three different kinds of seed, and that two of these : Individual Plant 

 Seed (o), and Individual Band Seed {ß) were selected, and further the amounts 

 of each subgroup sown in tlie different crops wonKI naturally differ rather widely. 

 Hence the niean and variability of any crop eannot be effeetively compareil with 

 that of the first Hampden crop oii the ba.sis of either (a) or {ß). For coniparison 

 the General Seed (7) was provided, but unluckily this failed in two instances; 

 at Bookham no return for (7) wiis made, and at Highgate the experimental patch 

 of (7) produced only one or two starveling plants. We have the followiug 

 results : 



* MessrB Sutton & Co. to whom I am indebted for Information on this point hold that the relative 

 ezlent depcnds on the etate of weathcr at the time of flowering and on other conditions. 



