SYNOPSIS, The nineteenth century was critical for the empirical and co
nceptual growth of developmental biology, Fishes played a central role
in this process, The study of fish development, mainly that of teleos
ts but also chondrichthyans, can be traced back to classical times, In
the nineteenth century, it merged with modern descriptive embryology,
continued with the rise of comparative embryology associated with evo
lutionary studies, and moved into the experimental and physiological a
nalysis of development, Any consideration of fish development must tak
e into account that fishes phylogenetically are the most diverse group
of the vertebrates and also the most speciose, These features are ref
lected in the diversity of their development, The descriptive embryolo
gy of fishes is reviewed from Aristotle to the beginning of the ninete
enth century. The study of chondrichthyans, especially viviparous spec
ies, was characteristic of this period, During the nineteenth century,
there was a progressive development of knowledge of the descriptive e
mbryology of teleosts and chondrichthyans. Teleosts came to the fore b
ecause artificial fertilization ensured a ready supply of material and
their transparent eggs were well suited for microscopy, The subsequen
t development of embryological microtechnique made possible the examin
ation of sectioned material and moved research to a more cellular leve
l. By the end of the century, an in-depth description of development w
as in place, Interest in the comparative embryology of fishes was stim
ulated by Haeckel's melding of embryology and evolution and led to a d
escription of development of agnaths, chimaeras, lungfish, and primiti
ve actinopterygian fishes, Experimental and analytical methods of inqu
iry began to be used at mid-century, The experiments of Ransom on the
contractility of egg cytoplasm, Lereboullet's experimental teratology,
chemical studies of embryonic nutrition in viviparous fishes, in vitr
o observation of blastomeres, His's concrescence theory of embryo form
ation and Kastschenko's and Morgan's testing of it are considered.