STUDIES ON EXPERIMENTAL HAPLOIDY IN SALAMANDER LARVAE

II. CYTOLOGICAL STUDIES ON ANDROGENETIC EGGS OF TRITURUS VIRIDESCENS

¹ From the Department of Anatomy, Medical College, Syracuse University and the Marine Biological Laboratory, Woods Hole, Massachusetts

1. Androgenetic eggs of Triturus viridescens most frequently cease development in the following stages : a. Irregular beginning cleavage; b. Early blastula; c. Late blastula; d. Gastrula.

2. The causes of arrested development were investigated cytologically in eggs fixed in each of these stages.

3. Eggs of the first group were of two types, i.e., abortive cleavage, and early irregular cleavage in which a few cells were formed near the animal pole. In seven eggs of the first type, it was found that the sperm nuclei had degenerated either before or during early mitosis and cleavage furrows had disappeared. In five eggs of the second type, either all sperm nuclei had degenerated during early mitosis or one sperm nucleus divided more or less normally while "accessory" sperm nuclei either degenerated or divided irregularly in the unsegmented part of the egg.

4. In nineteen early blastulae, thirteen were incompletely segmented and six, although irregularly segmented, were fairly normal blastulae. Associated with these abnormalities in the thirteen eggs were the independent division of sperm nuclei in the yolk region without segmentation of the cytoplasm, and the presence of abnormal mitoses in the majority of cells. In the six almost normal mid-blastulae, the greater number of cells contained abnormal nuclei. Chromosome counts varied from 9 to 22 + in twelve of the nineteen eggs in which analyses could be made.

5. In twenty-three late blastulae sectioned, the same abnormalities as found in the earlier blastulae were observed. The majority of eggs were incompletely segmented and all of the eggs contained abnormal mitotic figures in some of the cells. Chromosome counts were made in fourteen eggs. None of these blastulae were completely haploid.

6. Nine gastrulae examined were abortive. No abnormal mitotic figures were found in these eggs, but in seven gastrulae the chromosome numbers varied above and below the haploid number, indicating that abnormal mitoses had occurred during earlier cleavage stages. Two gastrulae were haploid and it is assumed that these are examples of the abnormalities which many haploid embryos exhibit when differentiation of parts or of structures first takes place.

7. These observations confirm and extend those of Fankhauser and of Fankhauser and Moore. In order to develop beyond the gastrula stage, an androgenetic egg must be at least completely hapboid.