March 31, 1922] 



SCIENCE 



349 



different size and shape and possesses import- 

 ant differences in nearly all parts of the plant, 

 the discrimination against the foreign pollen is 

 very pronounced. The result is almost com- 

 plete non-functioning of the pollen from the 

 dissimilar plants although such pollen when 

 acting alone is capable of normal fertilization. 

 The greater genetic diversity of these two types 

 is also indicated by the fact that the amount of 

 heterosis shown in the increased weight of the 

 crossed seeds is much more than in the previous 

 case of the similar varieties. 



There is here exhibited the working of a ten- 

 dency which acts to set individuals apart. Be- 

 sides Zea mays three other species, representa- 

 tives of different orders of the two main classes 

 of angiosperms, show the same phenomenon. 

 It is paralleled in the assortative mating of 

 animals from the lowest to the highest. It is 

 not inconceivable that when carried far enough 

 there may be created an impassable physiolog- 

 ical barrier separating different groups. As 

 far as can be judged the differences shown by 

 the types used for illustration are the usual 

 qualitative and quantitative characters which 

 we have been thinking of in terms of Men- 

 delian units. Such hereditary characters may 

 not be directly concerned with the selective 

 action but may be merely associated with dif- 

 ferences in more fundamental qualities, but 

 whatever these are they are transferable. 



Of course this tells us nothing as to how 

 the diff'erences which are correlated with in- 

 equality in fertilizing ability arose. But how- 

 ever diverse were the forms which entered into 

 the ancestrj' of maize they must have been 

 sexually compatible. Individual members of 

 this species which are quite diverse in form 

 and behavior are now showing a marked ten- 

 dency towards sexual incompatibility. The 

 degree of selectiveness may be no greater now 

 than it was when the species was first founded 

 but the fact that there is a condition of measur- 

 able physiological aloofness is reason to sup- 

 pose that the accumulations of characters of 

 the same order would culminate in different 

 groups being clearly set apart. Given sufficient 

 time specific differences may finally result. 



D. F. Jones 

 Connecticut Agricultural 

 Experiment Station 



GRAVITATIONAL ABSORPTION 



The experiments of Majorana^ on gi-avita- 

 tional absorption having attracted considerable 

 notice, it seems well to direct attention to the 

 large amount of experimental evidence to the 

 contrary. 



Russell- in a recent article has shown that 

 astronomical and tidal phenomena would limit 

 any gravitative absorption to one-thousandth 

 or less of the amount announced by Majorana. 



Eichelberger and Morgan,^ of the U. S. Naval 

 Obseiwatory, have recently published the re- 

 sults of clock observations from 1903 to 1911, 

 reduced so as to show a dift'erenee, if any, be- 

 tween the day and night rate. It appears from 

 these results that such a difference can not ex- 

 ceed 0.005 second, about one-tenth that pre- 

 viously announced by the Lick Observatory*. 

 The writer is verbally informed that the Naval 

 Observatory results from 1913 to 1918, reduced 

 but not yet published, bring this slight dif- 

 ference to a stiU smaller figure. 



The existence of gravitative absorption 

 should cause a pendulum clock to run slower 

 by night than by day, on account of the ab- 

 sorption by the earth of the gi-avitative action 

 between the pendulum bob and the svm. Tak- 

 ing Majorana's coefficient of absorption 

 (6.7 X 10—12) the average absoi'ption of gi-avi- 

 tation by the earth during the night would 

 amount to about three per cent, of the solar 

 gravitative acceleration at the distance of the 

 earth, which is about 0.0006g. The total 

 gTavitative acceleration to which the pendulum 

 bob is subjected would therefore be reduced 

 at night by about 0.00002g, or 2 parts in 

 100,000, and the average time of swing in- 

 creased by 1 part in 100,000. 



Taking 0.005 second as the greatest permis- 

 sible change in 12 hovu-s, the observations of 

 Eichelberger ' and Morgan would limit the 

 change in time of swing to something like 1 

 part in nine million. 



Majorana himself, in a later scries of expeii- 



1 FUil Mag., 39: 488, 1920. Atti della Beale 

 Aecademia dei Lincei, 28, 1919, aud 29, 1920. 



~ Astrophysical Journal, 54 : 334, 1921. 



3 Astronomical Journal, No. 795, January 1922. 



* Lick Observatory Bulletin, No. 330, April, 

 1921. 



