280 



SCIENCE 



[N. S. Vol. XXXIV. No. 870 



The results of the work are best typified in 

 a single group of experiments which were 

 made on May 21. Twelve embryos of Bana 

 palustris were used and from each the whole 

 neural tube with the adjacent mesoderm was 

 cut out in a dish of saline. In each ease the 

 piece of tissue was divided into two parts and 

 one half mounted in a simple hanging drop 

 of defibrinated frog's serum, the other half in 

 a drop of the same fluid held between two 

 layers of spider web, as described above. 



Of the twelve specimens in the first group, 

 none showed any active cell movements dur- 

 ing the first six days, and after that only one 

 specimen, to be referred to below, mani- 

 fested anything of the kind. In these speci- 

 mens many cells became loosened from the 

 main mass of tissue, remaining inactive and 

 rounded, though the main mass ilfself re- 

 mained alive and in a number of cases dilfei}- 

 entiated into striated muscle which e'jdbifeited 

 frequent twitching. 



Of the twelve specimens mounted between 

 webs, eleven showed very active movements of 

 the embryonic cells, which began even on the 

 same day on which the preparation was made. 

 Only one of these preparations gave negative 

 results, and this was one that was injured by 

 rough handling in mounting. The behavior 

 of the cells in these cases is fundamentally 

 different from that of the other cells in the 

 simple drop without support, and the general 

 appearance of the preparation is not unlike 

 that of specimens isolated in clotted lymph, 

 though there are some differences. Numerous 

 cells extend from the main mass of tissue, 

 sometimes singly, sometimes in masses. The 

 cells are spindle shaped, branched or poly- 

 gonal, with hyaline protoplasm in the proc- 

 esses and at the angles of the cells, the cell 

 body being gorged with yolk granules. Care- 

 ful focusing shows that the active cells are 

 confined to two levels, viz., the under surface 

 of the cover slip and the lower surface of the 

 drop which rests on the layer of web spanned 

 across the glass ring. Cells which are partly 

 loosened between these two levels remain 

 rounded and inactive, just as in the hanging 



drop preparations. In a very large number of 

 cases the active cells are found to have definite 

 relations to the web fibers. Often strands of 

 spindle-shaped cells, resembling the cells of an 

 embryonic tendon or the Schwann cells of a 

 developing nerve, are found closely applied to 

 slender bundles of web fibers. Again, spindle- 

 shaped cells, sometimes with very long proc- 

 esses, lie in intimate contact with single web 

 fibers. Where two such fibers cross, the cells 

 may assume a tri- or quadripolar shape, with 

 a process running along each fiber. Fre- 

 quently the cells are closely attached to the 

 cover slip and are then usually of flattened 

 polygonal shape, forming in many cases ex- 

 tensive sheets. The cells may change their 

 shape and move from place to place, or they 

 may remain in one spot for days, practically 

 unaltered in shape. After a few days typical 

 pigment cells developed in a number of cases 

 and these too assumed definite relations to 

 the web fibers. In only two of the specimens 

 were outgrowing nerve fibers observed. They 

 were in all essentials like those previously 

 found in the clotted lymph preparations, and 

 in each case they crept along the lower surface 

 of the cover slip, without definite relations to 

 the web. 



In this series of experiments the contrast 

 between the preparations in the free hanging 

 drop and those supported by web is so marked 

 and so constant that it is impossible to escape 

 the conclusion that the cells are able to execute 

 their movements only when some solid frame- 

 work is given them. The experiments in 

 which saline solutions were used instead of 

 serum bear out this same conclusion, though 

 they are not so striking on account of a larger 

 proportion of them giving negative results. 

 The saline solutions are clearly not such good 

 culture media as the serum, but notwithstand- 

 ing this a number of specimens showed 

 marked cell movement and remained alive for 

 days. 



Some of the experiments at first sight seemed 

 to oppose the conclusion stated above, but on 

 more careful study were found to afford a 

 striking confirmation of it. This was brought 

 out especially in a series cultivated in Locke's 



