June 30, 1905.] 



SCIENCE. 



ingly shows one more than one half the sperma- 

 togonia! number of chromosomes, but at the 

 close of this division conjugate to form an 

 asymmetrical dyad, the number of separate 

 chromatin-elements being thus reduced from 

 eight to seven (in Podisus from nine to eight). 

 A reduction of the number to seven in the 

 first division, such as has been described by 

 Montgomery as an occasional or usual process 

 in Euchistus and Coenus, was never observed. 

 In the second division the asymmetrical idio- 

 chromosome-dyad separates into its unequal 

 constituents, while the other dyads divide sym- 

 metrically. One half the spermatozoa, there- 

 fore, receive the large idiochromosome and one 

 half the small, the other chromosomes being 

 exactly duplicated in both. 



Correlated with this asymmetry of distribu- 

 tion is the fact that the spermatogonial chro- 

 mosome-groups do not show two equal micro- 

 chromosomes (as is the case in such forms as 

 Anasa, Alydm or Protenor, where an accessory 

 chromosome is present) but only one, which is 

 obviously the small idiochromosome, the large 

 one not being certainly distinguishable at this 

 period from the other spermatogonial chromo- 

 somes. The final synapsis of the idiochromo- 

 somes is deferred to the prophases of the second 

 division, somewhat as that of the two equal 

 microchromosomes is deferred until the pro- 

 phase of the first division in Anasa, Alydus 

 and some other forms. A remarkable result 

 of the diiference in this regard between the 

 forms that possess and those that lack a true 

 accessory chromosome is that in the former 

 case {Anasa, Alydus, etc.) the first division of 

 the small central chromosome is a reduction- 

 division and the second an equation-division; 

 while in the latter case {Lygwus, Cainus, etc.) 

 the reverse order manifestly occurs. The rela- 

 tion of these observations to earlier ones by 

 Paulmier, Montgomery and others was pointed 

 out, with a discussion of their bearing on the 

 Mendelian phenomena of heredity and the 

 problem of sex-determination. 



Professor Crampton presented briefly some 

 of the conclusions drawn from the results of 

 his work upon variation, correlation and selec- 

 tion among saturnid lepidoptera. The earliest 



studies showed that eliminated individuals, 

 when compared with similar members of the 

 same group that survive, prove to be more 

 variable and of somewhat different types, 

 although this relation between variability and 

 selection is not a constant one. The charac- 

 ters utilized for these preliminary studies, 

 namely, certain pupal dimensions and propor- 

 tions were of such a nature that they could not 

 serve the pupa directly in any functional man- 

 ner, wherefore it was concluded that their con- 

 dition of correlation formed the actual basis 

 for the selective process, formative correla- 

 tion being also distinguished from functional 

 correlation. That the general condition of 

 correlation among the structural characters of 

 pupte formed, indeed, the basis for selection 

 was further indicated by the results of a sta- 

 tistical study of the correlations between vari- 

 ous characteristics of pupal groups from several 

 different animal series ; although an advantage 

 did not always appear in favor of the surviving 

 group. On the basis of the foregoing, a gen- 

 eral theoretical conception was developed, ac- 

 cording to whith the whole series of internal 

 elements and the whole series of external in- 

 fluences were regarded as involved in the 

 determination of the general condition of cor- 

 relation or coordination that formed the basis 

 for selection, as adaptive or the reverse. 



M. A. BiGELOW, 



Secretary. 



DISCUSSION AND CORRESPONDENCE. 



PRE-PLEISTOCENE DEPOSITS AT THIRD CLIFF, 

 MASSACHUSETTS. 



To THE Editor of Science: It has been 

 suggested by several writers (Shaler and Ver- 

 rill) that Tertiary and Cretaceous deposits 

 may occur on the floor of the sea north of 

 their known occurrence on Marthas Vineyard 

 and Cape Cod. Their northerly occurrence 

 on land has not been noted except for the 

 Miocene greensands at Marshfield, Mass. 

 (Duxbury sheet, U. S. G. S.). During the 

 spring field season at Harvard University the 

 writer reexamined the coast from Boston 

 Harbor to Peaked Cliff, fifteen miles south- 

 east of Plymouth harbor, in order to test, by 



