694 THE RESPIRATION AND [pt. iii 



freshly laid eggs showed them that an appreciable amount of carbon 

 dioxide could be given off from bicarbonates in the shell — a finding 

 that might have been expected in view of the fact that the oviduct 

 of the hen is probably saturated with carbon dioxide, unlike the 

 air outside. Compare with this result the work of Bialascewicz & 

 Bledovski on amphibian eggs^. These experiments with shells alone 

 provided Bohr & Hasselbalch with a correction which they intro- 

 duced into the values obtained for fertile normally developing eggs. 

 This correction, while negligible in later stages of development, was 

 very important in the early stages where the respiration of the embryo 

 is small. More often they got over the difficulty by keeping fertile 

 eggs at room temperature in a current of carbon-dioxide-free air 

 until they gave ojff no more of the gas. Controls carried through for 

 3 weeks on infertile eggs showed that Pott & Preyer had been wrong 

 in their conclusion that infertile eggs gave off notable quantities 

 of carbon dioxide, for, on the contrary, the amount given off was 

 remarkably small, varying from o to 5 mgm. per 24 hours. Then they 

 proceeded to the experiments with the developing embryos, obtaining 

 the diagram shown in Fig. 140. The varying width of the columns 

 is a measure of the length of time the experiment was conducted ; 

 thus in one or two cases, it was as much as 24 hours, but in the 

 majority only 4 or 5. A glance at the graph shows the initial pro- 

 duction of gas from the shell, the steady rise from the 2nd day 

 onwards, and the plateau which Bohr & Hasselbalch always got 

 from the 17th to the 21st day. Their next interest was the variation 

 in the respiration intensity. For this it was necessary to make 

 weighings of the embryos, for at that time the fragmentary values 

 of Falck were all that were available. The numerical results 

 which Bohr & Hasselbalch obtained are shown in Appendix i, 

 and agree well enough with those got by later observers, but they 

 made the interesting correlation that the curve for total weight 

 (not weight increments) went up in exactly the same manner 

 as the curve for the carbon dioxide excretion in cubic centimetres per 

 24 hours (i.e. increments of respiration) . This is shown in Fig. 141 

 taken from their paper, and it should be noted that the agree- 

 ment is rather better after the 9th day than it is before it. The 

 metabohc rate values are shown in Fig. 142 constructed from their 

 data. A sharp descent brings the metabolic rate down to what is 



