144 CARNEGIE INSTITUTION OF WASHINGTON. 



numerically low, ranging from 0.023 to 0.355, with an average value 

 of 0.195. For pods per plant and seeds per pod the correlations are 

 also for the most part positive. The values range from —0.046 to 

 +0.338 with a mean of +0.126. By all available tests the correlations 

 for pods per plant and number of ovules per pod seem to be signifi- 

 cantly higher than those for pods per plant and number of seeds per 

 pod. Analysis shows, however, that the relationship between the size 

 of the plant and the number of seeds per pod is to some degree inde- 

 pendent of that for number of pods per plant and number of ovules 

 formed per pod. 



THE PRODUCTION OF AN IMPROVED VARIETY OF SHEEP BY 

 MODERN METHODS. 



This research is carried on primarily by the New Hampshire Experi- 

 ment Station with the collaboration of the Director. By crossing a 

 mutton and a fine-wool type of sheep, and by judicious selection, it is 

 hoped to secure a strain of sheep that will combine the most desirable 

 characteristics. Already in the F2 generation very valuable combina- 

 tions of characters have appeared; but it is necessary to breed great 

 numbers to get the desired quality, and by appropriate breeding to 

 fix the desired combination of traits. 



ORIGIN OF THE CHARACTERISTICS OF CAVE ANIMALS. 



Selection experiments with Cladocera. — These investigations of the 

 problem of the origin of the light-shunning reactions of cave animals 

 give promise of affording further evidence of the vahdity of Johannsen's 

 hypothesis. The mass of data is large, is being carefully compiled, 

 and will undoubtedly constitute a sound piece of evidence bearing 

 on a hypothesis the universal applicability of which is by no means 

 demonstrated. 



Additional Cladocera material. — Four lines of an additional species 

 of Moina, of the order Cladocera, are now being kept under control 

 conditions similar to those of the other species mentioned. All the 

 lines of the four species of Cladocera now being bred parthenogeneti- 

 cally are being used in some phase of the experiments regarding the 

 supposed sexual cycle and also in experiments on heredit}^ in partheno- 

 genetic reproduction. 



Experiments on the cave problem. — Dr. Banta's experiments upon the 

 effect of light on true cave species and of a cave envii'onment upon 

 outside forms are progressing favorably' and as rapidl}' as the nature of 

 the problem permits. Some additional cave species are being reared 

 in daylight and some additional outside species in a cave environment. 

 The second generation of the amphipod Eucrangonyx gracilis, derived 

 from the non-cavernicolous race of this species, reared in the cave, 

 is not yet large enough to determine how much, if anj', pigment will be 

 produced. The first-generation offspring derived from the surface 

 stream race of this amphipod kept in the cave show in some individ- 



