THE MECHANISM OF PHOTOSYNTHESIS 301 



the grana may be submicroscopic or may exhibit uo optically visible 

 phase boundaries and that they may become visible under certain 

 conditions by "coarsening," which he does not consider as due to 

 artifacts. 



Electron-microscope studies have indicated that the grana are truly 

 morphological units. According to Granick (1949), a mature spinach 

 chloroplast contains 40-60 grana, about 6000 A in diameter and 800 A 

 thick, embedded in the protein-containing matrix. He holds that the 

 protein remaining after extraction of the grana with methanol makes up 

 less than half the original material. Recently, Thomas et al. (1952) also 

 made an electron-microscope study of spinach chloroplasts and grana. 

 They confirmed the earlier findings of a membrane around the chloroplast, 

 and by applying both pepsin and lipase digestion they demonstrated that 

 this membrane, Hke the stroma, contains both proteins and lipids. The 

 proteinaceous framework in which the grana are suspended shows a 

 "thread-globule" structure. The globules contain substances that have 

 a larger electron-scattering power than those found in the threads. After 

 digestion of the proteins a spongy lipoid mass remains. According to 

 these authors the grana are also surrounded by a protein-lipid membrane. 

 They confirm observations of Frey-Wyssling and Aiiihlethaler (1949) that 

 grana consist of protein discs separated by lipids. 



The renewed study of the grana has led to the view that they are the 

 only seat of the chlorophyll. Evidence for this is contained in an obser- 

 vation by -lungers and Doutreligne (1943), who observed green grana 

 and a white matrix against the background of starch grains in amyloplasts 

 of potato exposed to light. Metzner (1937) succeeded in demonstrating 

 chlorophyll fluorescence of grana while the stroma remained dark. 



The available evidence suggests a structure of the granum of the type 

 already proposed by Hubert in 1935 for the entire chloroplast. Accord- 

 ing to this suggestion, hydrophilic protein layers alternate with lipophilic 

 layers, in which the pigments and lipids, e.g., cholesterol molecules, occur 

 in a definite arrangement and have a definite orientation with respect to 

 the protein. Frey-Wyssling (1948) estimates that a granum contains 

 20-30 parallel fatty layers, each having a thickness of about 50 A, sepa- 

 rated by aqueous protein layers having thicknesses of about 250 A. 



The view has developed that, in the chloroplasts, chlorophyll is bound 

 to protein. An important argument for this was derived from spectrum 

 studies which showed that in the living cell the red absorption maximum 

 was at a longer wave length than in any solvent. This was much more 

 so in the bacteriochlorophyll of the purple bacteria, where the difference 

 may be as much as 100 m/x (Katz and Wassink, 1939). Extraction of 

 bacteriochlorophyll from various strains yields solutions with an absorp- 

 tion maximum at approximately 770 vajx in ethanol, whereas the maxima 

 in the living bacteria are found at various places between 800 and 900 mu 



