The Scale of Structural Units in Biopoesis 397 



form implies a more complex folding or coiled coiling of the polypeptide chain 

 than occurs in simple spirals. What precisely is this configuration has yet to be 

 determined in any single case. This is the great object of protein crystal analysis 

 but it must be admitted that so far this has led only to negative results : that is, 

 we know that some of the simpler arrangements of parallel-packed spirals cannot 

 account for the observed intensities of X-ray diffraction from globular protein 

 crystals. We cannot reaUy claim, though here success may be very close, that 

 even the principle of packing of the chains in globular protein is yet understood. 

 All that can be said is that there appears, from the study of the distribution of 

 the shorter interatomic vectors in the analysis of their crystals, that more than 

 one basic arrangement is possible. The configuration in ribonuclease [i6, 17] 

 differs from that in haemoglobin and related compounds [18, 19]. 



However, recent work by Kendrew [31] on the structure of whale myoglobin, 

 using multiple heavy-atom substitution to fix the complex phases of X-ray 

 reflections, indicates an irregularly bent tangle, presumably of polypeptide hehces 

 surroimding the haem group. 



The biological significance of the folding of protein chains to form globular 

 molecules is not altogether clear. It would appear from studies on enzymic 

 activity [20] that this is equally effective in the extended form produced in 

 strong urea solutions. Further, from the amount of space available, about 10- 

 20 Â, presumably, for enzyme protein molecules in the interstices of mito- 

 chondria, they can only be in the form of single or double extended peptide 

 chains. Nevertheless, in solution in tissue fluids and possibly also in intracellular 

 fluid, they are present as compact thicker molecules. Also, protein crystals, 

 presumably formed of globular molecules, are occasionally found in cells whüe 

 the outer shells of viruses and possibly of microsomes as well are also formed of 

 them. We may provisionally, therefore, think of the globular form as one state, 

 possibly only the equilibrium or resting state, of the protein molecules. The 

 biological importance of the larger globular molecules may indeed be rather 

 physico-chemical and colloidal. They would secure by their shape the maximum 

 of easy diffusion and by their size a relative freedom from thermal disturbance. 

 This would give them an importance in maintaining osmotic pressure and as a 

 transport mechanism for amino acids. 



Evidence is accumulating that the larger globular protein molecules of mol. wt 

 above 14,000 and even less, as in the case of insuHn, are formed of secondary 

 agglomerations of identical or quasi-identical subunits of the minimal simple 

 peptide of between 30 and 100 monomers. This can be proved only by reversible 

 disaggregation, as in the case of haemoglobin with four imits which can be split 

 first into two and then into four subunits of the myoglobin type. The way this 

 disaggregation is affected, by the use of extreme pH or strong ionic conditions, 

 suggests that the links between the units are ionic and a similar explanation would 

 accoimt for the triple association of zinc insulin. With even bigger molecules 

 such as haemocyanin, with a molecular weight of second or even third order, 

 agglomeration of an even more labile character seems to occur and here the sub- 

 units are large enough to be seen in the electron microscope. 



Another type of agglomeration of protein molecules seems to occur in crystal- 



