ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 
510 
see how far the chondrocranium can he employed as an aid in classifica- 
tion. Thus he finds evidence against associating the Caecilians with any 
of the Urodela, and in favour of keeping them in a distinct group co- 
ordinate with the Urodela and Anura. As regards the Dipnoi, the 
evidence confirms W. N. Parker’s view that they should not be retained 
among the fishes. 
Early Stages in the Development of Salmonidae.* * * § — Dr. J. Jablo- 
nowski finds (in the salmon) that the endoderm, first established by the 
gastrulation, forms the definitive lining of the most anterior blind cupola 
of the archenteron, which afterwards undergoes degeneration ; that the 
anterior boundary of the rudiment of the central nervous system corre- 
sponds with the roof of this archenteric cupola ; and that the dorsal 
blastoporal lip is originally situated not far behind this anterior boundary. 
In the latter part of his paper, he works out four points of agreement 
between the trochophore and the vertebrate embryo. 
Development of Teleostean Scales.! — Herr S. A. Ussow has studied 
this in Cobitis tsenia , Ac. The upper layer consists of a structureless 
homogeneous material formed, at the expense of the mesoderm of the 
cutis, from the scleroblasts. Its inorganic substance is amorphous cal- 
cium phosphate. In short, the tissue is a simple form of bone. The 
lower layer is formed from the same mesodermic elements, and consists 
of hardened connective tissue threads without any cells between. 
Development of Caecum in Amphioxus.ij: — Prof. J. A. Hammar finds 
that this organ arises as a median “ liver-fold,” which is constricted off 
caudalwards to form a blind sack. Thus the development is like that of 
a typical vertebrate liver. The position on the right side is secondary. 
The chief peculiarity is the non-formation of any parenchyma. 
Law of Ancestral Heredity.§ — Prof. K. Pearson expounds and illus- 
trates what he terms Gallon’s Laiv of Ancestral Heredity , which “ enables 
us to predict a priori the values of all the correlation coefficients of 
heredity, and forms the fundamental principle of heredity from which 
all the numerical data of inheritance can in future be deduced, at any 
rate, to a first approximation.” 
In the notation adopted, a father is called a 1st parent, a grandfather 
a 2nd parent, a great-grandfather a 3rd parent, and so on ; the mid s ,h 
parent or the s ,h mid-parent is derived from all 2" individual s th parents. 
With reservations as to how “ mid-parent ” should be defined, Pearson 
states the law of ancestral heredity as follows. 
“ If & be the deviation of the s ,h mid-parent from the mean of the 
s ,h ancestral generation, and h 0 be the probable deviation from the mean 
of the offspring of any individual, o- s the standard deviation of the s th mid- 
parental generation, cr 0 of the generation of the offspring, then 
7 1 U'o 7 | 1 ®"o 7 t 1 U"o 7 . 1 U’o 7 . 
^0 — 9 — ^1 4- t — *2 4- q — + Ya — *4 + 
2 CT l 4 <T 9 O O-o lb (Ta 
* Anat. Anzeig., xiv. (1898) pp. 532-51 (19 figs.), 
t Bull. Soc. Imp. Nat. Moscou, 1897 (published 1898), pp. 339-51 (2 pis.). 
X Anat. Anzeig., xiv. pp. 602-7 (5 figs.). 
§ Proc. Boy. Soc., lxii. (1898) pp. 386-412. 
