494 
L. DONCASTER. 
l^ie second piece of direct evidence connecting chromo- 
somes with inherited characters is due to Baltzer (5). 
Previous investigators had found that the cross between the 
sea-urchins S p h as r e c h i n u s g r a n u 1 a r i s $ and Strongylo- 
centrotus lividus B produces plutei intermediate in 
character between those of the parent forms^ while the con- 
verse cross, in the rare cases when the larvae survive, gives 
plntei in which the skeleton is of the pure maternal form. 
Baltzer has examined the behaviour of the chromosomes in 
the two crosses. He finds no peculiarity in the cross with 
S p h ae r e c h i n u s $ , but when Strongylocentrotus is used 
as the female parent, he finds that about 16 chromosomes 
are constantly omitted from the daughter-nuclei at the first 
and second segmentation division of the egg. He infers by 
three distinct methods that the eliminated chromosomes are 
paternal, i.e. derived from Sph ae rechi nu s, and since the 
haploid number of Sphterechin us is 20, there remain 
only about four Sphmrechinus chromosomes in the hybrid 
plutei, which are maternal in character, while the full number 
remains in the intermediate plutei produced by the converse 
cross. That the eliminated chromosomes are paternal is 
inferred (1) from the shapes and sizes of the chromosomes ; 
(2) from a study of a tetraster, in which the number corre- 
sponds with the expectation if two spermatozoa had entered 
the egg and only eight of the sperm-chromosomes behaved 
normally ; (3) by the fertilisation of non-nucleated St rongy- 
locentrotus egg-fragments with Sphasrechin us sperm, 
in which again only about four chromosomes divide normally. 
Tenuent (60) similarly finds elimination of chromosomes in 
the cross Hipponoe ? x Toxopneustes c? , but none in the 
converse cross. In this case the Hipponoe characters 
dominate in the pluteus whichever way the cross is made, 
but since the Hipponoe chromosomes are presumably present 
liirtaria and Nyssia zonaria. In this case the hybrids are sterile, so 
the later generations are not available. The work is to be published in 
the next number of the ‘Journal of Genetics.’ 
