SPECTRA OF PHOTOSYNTHETIC PIGMENTS 385 



the 590-m/x bacteriochlorophyll peak. It is clear that the spirilloxanthin 

 that causes most of the carotenoid absorption in Rhodo spirillum ruhrum 

 is not the active carotenoid, since the absorption bands and the action 

 peaks do not agree. These two action spectra do, however, rather closely 

 match the action spectra for phototaxis and for fluorescence excitation of 

 Chromatium shown in Fig. 6-156. In Chromatiuni the main carotenoid 

 absorption peaks match those of the action spectra in position but not in 

 height. 



Transfer of Energy from Bacterial Carotenoids to Bacteriochlorophyll 

 as a Step in Photosynthesis. We have seen in Figs. Q-V2d and 15a and c 

 that photosynthesis and phototaxis in purple bacteria are carried out 

 not only by light absorbed by the bacteriochlorophyll itself but also by 

 light absorbed by some of the carotenoids that are present in small 

 amounts. That the energy absorbed by carotenoids is transferred to 

 bacteriochlorophyll is indicated by the close match between the action 

 spectrum for the excitation of bacteriochlorophyll fluorescence in Fig. 

 6-15c and that for phototaxis in the same bacteria, Chromatium. Quan- 

 titative measurements of fluorescence in Rhodo spirillum rubrum, also 

 by Duysens (personal communication, 1950), indicate that the carotenoid 

 activity in this species also takes place by transfer of energy to the 

 bacteriochlorophyll. 



In addition to energy transfer from carotenoids to bacteriochlorophyll, 

 a transfer from one type of bacteriochlorophyll molecule to another has 

 b'een found by Duysens. This may be analogous to the transfer from 

 chlorophyll b to a in green plants (Duysens, 1951a) and in solutions 

 (Watson and Livingston, 1950). In Fig. 6-13a and c are presented the 

 fluorescence spectra and the near-infrared absorption spectra of Rhodo- 

 spirillum and of Chromatium. 



In Rhodospirillum one type of bacteriochlorophyll occurs with an 

 absorption peak at 890 m/x; in Chromatium three types occur which have 

 absorption maxima at 800, 855, and 890 m^i. The presence in Chromat- 

 ium of the type with the longest-wave-length absorption peak is indicated 

 by the hump at 890 m/i in the absorption spectrum. Quoting Duysens 

 (1951a): "The fluorescence spectra of Chromatium and Rhodospirillum 

 rubrum are similar, so probably also the absorption spectra of the emitting 

 molecules are similar. The absorption spectrum of Chromatium might 

 thus be anatyzed in three spectra similar to the absorption spectrum of 

 Rhodospirillum rubrum; only the molecules with absorption peak at 890 

 mix show fluorescence. Upon heating, these molecules disappear, as 

 shown by the absence of the hump at 890 m/x in the heated autolysate 

 [Fig. 6-156, d], and a new fluorescence peak occurs, probably from the 

 molecules with an absorption peak at 855 m^u. This experiment sug- 

 gests that in the presence of the molecules with an absorption peak at 890 

 m/i, transfer of energy occurs from the 855-m/i peak to the 890-m/i peak." 



