VIII] THE SEGMENTATION OF THE EGG 607 



patterns actually occur and in what proportions they do so in a 

 random sample of identical eggs. Some years ago Mr Martin 

 Adamson photographed more than a thousand frogs' eggs in my 

 laboratory, all at the stage shewing an eight-celled group of epiblastic 

 cells: with the remarkable result that every one of the twelve 

 possible arrangements was found to occur^ but some were common 

 and some rare, and the following were their comparative frequencies : 



In six separate batches of eggs (combined in the above list) one 

 or other of the first two types (c or^') was always the commonest; 

 and the first four taken together made up from 50 to 80 per cent, of 

 each separate sample. On the other hand, when Roux, many years 

 ago, shewed how various cell-configurations might be simulated by 

 oil-drops* — as we have done by means of soap-bubbles — he found 

 that the type i was essentially unstable, the large drop with its 

 seven contacts easily shpping into the centre of the system, and 

 there taking up a stable position of equihbrium. That the latter 

 is the more stable, and therefore the more probable, configuration, 

 seems obvious enough; and indeed type i seems so obviously 

 unstable that we are not surprised to find it at the bottom of 

 Martin Adamson"s fist of frequencies. The order in which the rest 

 occur is by no means so easy of explanation. 



There is a point worth considering in regard to the number of 

 contacts between cell and cell. In a system of eight cells, all 

 reaching the boundary and all with three-way junctions, there are, 

 besides the eight peripheral boundary-walls, thirteen internal parti- 

 tions, or 2 (n — 2) -|- 1 ; the number of interfacial contacts is double 

 that number, or twenty-six; and Ihe mean number of contacts for 

 each cell is 26/8, or 3-25. But, looking at the diagrams in Fig. 259 

 (which represent three out of our twelve possible arrangements of 



* Roux's experiments were performed with drops of paraffin suspended in 

 dilute alcohol, to which a little calcium acetate was added to form a soapy pellicle 

 over the drops and prevent them from reuniting with one another. 



