912 



SCIENCE 



[N. S. Vol. XXV. No. 649 



denies the transmission of unexpressed char- 

 acters through conjugate organisms, but in 

 dealing with gametes the distinction between 

 transmission and expression has continued to 

 be overlooked; otherwise the Mendelian hy- 

 pothesis of pure germ-cells could not have 

 attained its wide popularity. 



Mendelism and other forms of polarized ex- 

 pression inheritance yield us no intimation 

 whatever regarding the nature and mechanism 

 of transmission inheritance. If transmission 

 could be conceived as a matter of localized 

 character-unit particles we should be justified 

 in thinking of all germ-cells as containing 

 full sets, and not variously mangled fractions 

 of the ancestral equipments. Alternative in- 

 heritance of divergent characters means re- 

 ciprocal expression-polarities. It has yet to 

 be shown that there is any such phenom- 

 enon as alternative transmission inheritance, 

 brought about by the segregation of the pa- 

 rental character-units in different germ-cells. 

 Incompatibility suflacient to cause germinal 

 segregation should preserve the original asso- 

 ciation of the characters, but no such tendency 

 has appeared in Mendelian crosses. When 

 there are several divergent characters they are 

 always expressed in many different combina- 

 tions, as though to show that the scale of 

 transmission remains complete, no matter how 

 narrowly the needle of expression may some- 

 times be directed. O. F. Cook 



A NEW METHOD BY WHICH SPONGES MAY BE 

 ARTIFICIALLY REARED' 



I HAVE found in the course of an investiga- 

 tion carried on for the Bureau of Fisheries 

 that silicious sponges when kept in confine- 

 ment under proper conditions degenerate, giv- 

 ing rise to small masses of undifferentiated 

 tissue which in their turn are able to grow 

 and differentiate into perfect sponges. The 

 investigation has been prosecuted during the 

 past three summers at the Beaufort Labora- 

 tory. While the degeneration with the forma- 

 tion of the indifferent masses has been ob- 



' Published with the permission of Hon. Greo. 

 M.. Bowers, U. S. Commissioner of Fisheries, 

 served in several species, it is only in one 



species, a Stylotella, that the process as a 

 whole, has been worked out. 



This sponge, which is exceedingly abundant 

 in Beaufort Harbor, is a fleshy monactinellid 

 commonly reaching a thickness and height of 

 10-12 cm. Conical processes with terminal 

 oscula project upwards from the lower body. 

 With this species, which is a light-loving form, 

 I have obtained the best results when outside 

 aquaria, either concrete aquaria or tubs, were 

 used. The method of treatment is briefly this : 

 Into a tub about 60 cm. by 30 cm. and covered 

 with glass, a half dozen sponges, freed as far 

 as possible from live oysters and crabs, are 

 put. They are raised from the bottom on 

 bricks. The tub is emptied, filled and flushed 

 for some minutes three times in every twenty- 

 four hours. Direct rays of the sun should be 

 avoided. Tubs answer as well as concrete 

 aquaria, and have the advantage of being 

 movable. 



In a day or two the oseula of the sponge 

 disappear, and the surface begins to acquire 

 a peculiar smooth, dense and uniform appear- 

 ance. Microscopic study reveals the fact that 

 not only the oscula, but the pores also, for the 

 most part, close, and the canal system be- 

 comes interrupted and in some degree sup- 

 pressed. The mesenchyme is more uniform, 

 and is denser than in the normal sponge, ow- 

 ing in part at least to the disappearance of 

 the extensive collenchymatous (very watery 

 mesenchyme) tracts of the latter. 



The whole sponge may pass into this state 

 and remain without great change for weeks. 

 During this period it shrinks greatly in size, 

 in a given ease to one quarter the original 

 bulk. The arrangement of the skeletal spic- 

 ules becomes much simplified. With the 

 shrinkage in size the sponge becomes more 

 solid, i. e., more of the canal space is sup- 

 pressed. Some flagellated chambers persist 

 and there are a few small scattered apertures 

 on the surface. The bulk of the chambers dis- 

 appear as such, the collar-cells transforming 

 into simple polyhedral masses which become 

 scattered singly or in groups in the general 

 mesenchyme. The mesenchyme is a syncy- 

 tium composed of well-marked cells that are 



