kofoid: development of limax. 71 



(labelled c by Blochmann) also precedes in some instances in the later 

 divisions. The yolk, however, is confined mainly to two of the macro- 

 meres, A and />, instead of being lodged principally in three, A, B, and 

 C, as it is in Umbrella. 



All attempts to reconcile the cleavage of Chiton, as described by 

 Kowalevsky ('83) and Metcalf ('93), with the alternation of spirals 

 shown in other forms of Molluscan cleavage are involved in serious 

 difficulties. Unfortunately neither author figures nuclei or spindles, 

 with the exception of one figure in each paper. The relationship of 

 cells and the direction of the spiral is also explicitly stated in certain 

 cases to be a matter of conjecture on their part. Therefore it does not 

 seem profitable for me to add to their conjectures others of my own. 

 It is however possible to force upon the cleavage of Chiton, as figured 

 by these authors, an interpretation which causes it to accord with the 

 spiral type of cleavage, but this interpretation meets a serious obstacle 

 in the sixteen-cell stage, though it does not violate the relationships 

 of cells in the few cases where these authors have indicated relation- 

 ships by nuclear conditions (Kowalevsky, Plate I. Fig. 7, Metcalf, Plate 

 XV. Fig. xvii.). 



I do not Avish to commit myself to the view that Chiton conforms to 

 that type, for, as Metcalf has suggested, " most of the divisions are 

 of the radial type." The distribution of the yolk in the blastomercs is 

 also suggestive of the radial type. There are, however in Kowalevsky's 

 figures (Plate I. Figs. 4-13) many suggestions of spiral cleavage. In 

 Metcalf's figures, on the other hand, " individual variations in the shape 

 of the blastomeres are not preserved. The figures show what may be 

 called the typical condition." Thus most of the evidence of spiral 

 cleavage, if such exists, must be eliminated from his figures. 



The cleavage of Discocoelis conforms to the spiral type and the alter- 

 nation of spirals, with the possible exception of the spiral of 7.15-7.16, 

 to which reference was made in my former paper ('94, p. 19G). It is 

 necessary to relabel Figure 6 of Lang's Tafel XXXV., in order to reconcile 

 it with the apical views which he has given of the same stage, and also 

 with the principle of alternation. 



The conformity of the cleavage of Nereis to the law of altei'nation of 

 spirals is perfect, as is demonstrated by the application of the uniform 

 system of naming spirals in the above table. Even in those cleavages 

 designated as equatorial or meridional, traces of the spiral characteristic 

 of the generation can often be detected. 



The Nereis table is marked by the abrupt termination of the spiral 



