Planta (Berl.) 127,243--250 (1975) 9 by Springer-Verlag 1975
Chloroplast Fluorescence of C~ Plants I. D e t e c t i o n w i t h I n f r a r e d Color Film* L y n n e Elkin** and t~oderie B. Park Botany Department and Lawrence Berkeley Laboratory, Berkeley, California 94720, USA Received 21 April; accepted 30 July 1975
Summary. The precise spectral responses of infrared color film permit a rapid qualitative detection of the in situ fluorescence properties of chloroplasts in CAplants. Fluorescing mesophyll and granal bundle-sheath chloroplasts appear yellow, indicating a predominance of far-red over infrared fluorescence. In contrast, fluorescing agranal bundle-sheath chloroplasts appear red, indicating a predominance of infrared over far-red fluorescence. There is a strong positive correlation between lamellar appression and the ratio of far-red to infrared fluorescence as recorded by infrared color film. The red versus yellow color of bundle-sheath chloroplast infrared photographs is used to separate the 18 species of C4 plants and 2 species of Ca plants examined into two basic groups. This separation is compatible with a separation based on lamellar ultrastrueture of bundle-sheath chloroplasts. Introduction Over the past 10 years a n u m b e r of monoeots in the Gramineae and Cyperaeeae as well as representatives from at least 10 dieot families were shown to possess the Ca p a t h w a y of carbon fixation. Characteristically, C4 plants have low photorespiration and high photosynthetic capacity. I n addition, the leaves of these plants generally display " K r a n z " a n a t o m y in which the vascular bundles are surrounded b y bundle-sheath cells. The bundle-sheath cells m a y be agranal and generally contain larger chloroplasts t h a n the surrounding mesophyll cells (see Laetseh, 1974 for review). Investigations on C4 plants have been hampered b y the difficulty of isolating and completely separating intact bundle-sheath from mesophyll chloroplasts (Laetseh, 1971, 1974). Two of the more recent studies involved improved methods for bundle-sheath preparation (Andersen et al., 1971 ; E d w a r d s and Black, 1971 ; Mayne et al., 1971), although it is questionable whether anyone has yet isolated class i (Spencer and Unt, 1965) bundlesheath chloroplasts. Alternatively, fluorescence microscopy, an excellent high-resolution, nondestructive technique f o r monitoring the photochemical reactions of chloroplasts in ~itu, m a y be used to s t u d y these chloroplasts. The fluorescence pattern m a y be quickly surveyed visually, or recorded and measured photographically. Although the green colors of mesophyll and bundle-sheath chloroplasts in sugar-cane leaves * This work was supported in part by NSF grants GB-25579X and GB-41720X and the Atomic Energy Commission, and a Sigma Xi Grant-in-Aid of Research. ** Submitted in partial requirement of the Ph.D. degree. Lynne Elkin was supported as an NII-I Predoctoral Fellow. Present address: Department of Biological Sciences, California State University at Hayward, Hayward, California 94542, U.S.A.
244
L. Elkin and 1%. B. Park
appear equally i n t e n s e to the eye when viewed u n d e r bright-field microscopy, the large b u n d l e - s h e a t h chloroplasts disappear when viewed b y fluorescence microscopy. These b u n d l e - s h e a t h chloroplasts have either a drastically reduced fluorescence yield, or a p r e d o m i n a n c e of fluorescence in the infrared region of the spectrum to which the h u m a n eye is r e l a t i v e l y insensitive, or a c o m b i n a t i o n of both. I n f r a r e d color film, a film sensitive from 500-900 nm, is capable of distinguishing b e t w e e n visible far-red a n d infrared r a d i a t i o n ( E a s t m a n K o d a k Publication M-28, 1972). I n this paper we describe the m a r k e d difference in fluorescence p a t t e r n s of sugar-cane b u n d l e - s h e a t h chloroplasts, as well as those of other Ca plants, with the use of infrared fluorescence photomicroscopy. We explore the correlation b e t w e e n the degree of lamellar appression i n the chloroplasts a n d the relative proportion of far-red to infrared fluorescence as recorded b y infrared color film.
Materials and Methods Mature, dark-green and healthy leaves are freshly detached from Mollugo cerviana Ser., and Mollugo verticillata L. of the Aizoaceae, Amaranthus edulis Miehx. ex Moq. and Froelichia gracilis (Nook.) Moq. of the Amaranthaeeae, Atriplex lenti]ormis (Torr.) W0~ts. and Spinacia oleracea L. of the Chenopodiaccae, Euphorbia maculata L., Euphorbia serphylli/olia Pers., and Euphorbia splendens Bojer of the Euphorbiaceae, Cenchrus setigerus Steud. 1. e., Cynodon dactylon (L.) Pers., Dichanthium anuulatum (Forsk.) Stapf, Digitaria sanguinalcs (L.) Scop., Echinochloa colona (L.) Link., Saccharum o]/icinarum L., Sorghum ca//forum Be~uv. and Spartina [oliosa Trin. of the Gramineae, and Portulaca oleracea L. of the Portulacaceae. Except for Dig#aria, growth conditions do not seem to affect the fluorescence properties of the plants significantly since the results were indistinguishable whether plants were obtained from growth chambers, greenhouses, gardens, or collected from the field. In contrast, in Digitaria, the bundle-sheath chloroplasts varied from brilliant red when grown in greenhouses to pale orange when grown in growth chambers. Fresh leaf sections between 12 and 36 ~m thick are cut with a razor blade or with the aid of a Lab Line/Hooker (~[elrose Park, Illinois) plant Microtome No. 1225 and are placed directly into the buffers listed below. Chloroplasts are isolated according to the procedure of Jensen and Bassham (1966). They are suspended in Jensen-Bassham solution A, adjusted to pH 6.1 with KOtI; Jensen-Bassham solution C, adjusted to pH 7.2 with KOH; or solution E which consists of potassium-phosphate buffer, adjusted to p g 7.2 with KOH and containing the same inorganic ions as the other two buffers (Elkin, 1973). It is frequently necessary to remove excess starch grains and cell debris with a preliminary 2 rain 200 g centrifugation. Specimens are examined with a Zeiss fluorescence photomicroscope which incorporates high-pressure mercury vapor lamp HBO 200 W/4, ~ KGi heat-absorbing filter, a BG38 redsuppressing filter, a BG12 exciter filter, a heat-reflecting filter, the achromatic-aplanatie phase-contrast fluorescence condenser (N.A. 1.4), apoehromatie objective lenses and a No. 53 barrier filter. The condenser is coated with oil and set on its bright-field position to optimize the collection of light. The intensity of the fluorescence striking the film is maximized by mounting a camera back (without a lens) on top of the microscope in a direct line with the specimen and the objective lens to eliminate light losses in the automatic camera attachment. Eastman Kodak Infrared Aero Ektachrome 8443 is the infrared color film exclusively used in this investigation. The spectral response of each film emulsion is tested by photographing leaf sections from sugar-cane (a C~ plant) to assure uniform color response before a batch of film is accepted for experimental use. Since infrared color film is especially sensitive to emulsion deterioration, the film must be obtained fresh and immediately stored at -- 10~ until actual use. The film must be processed or refrozen (in the presence of a desiccant) immediately after it is exposed. For proper development, infrared color film requires strict Kodak E-3 processing (Eastman Kodak, 1972).
Chloroplast Fluorescence of C~ Plants I.
245
Results
A. Chloroplast Fluorescence Figs. l a and b show two patterns of color distribution of chloroplast fluorescence as recorded by infrared color film. All higher plants examined in this study display one or the other of these patterns which we have called Groups I and II. Group I plants are typified by the pattern shown in Fig. 1 a, an infrared color photograph of fluorescing chloroplasts in a leaf cross section of Dichanthium annulatum. The bundle -sheath chloroplasts appear red and the mesophyll chloroplasts appear yellow. Group I I plants are typified by the pattern shown in Fig. l b, an infrared color photograph of fluorescing chloroplasts in a cross section of a leaf of Spartina/oliosa. The photograph contains only yellow chloroplasts. Upon careful isolation, chloroplasts from both groups maintain their color distinction, as shown in Figs. 1 c and d. The red and yellow chloroplasts in Fig. 1 c were isolated from sugar-cane, a Group I plant, while the yellow chloroplasts in Fig. 1 d were isolated from spinach, a Group I I plant. The fluorescence properties of damaged chloroplasts are different from those of intact ones. Group I mesophyll and bundle-sheath chloroplasts suspended in H~O rupture osmotically and lose the clarity of the photographic red versus yellow distinction. Similarly, the color distinctions of very thin sections suspended in H20 deteriorate within minutes, in contrast to thin sections immersed in Jensen-Bassham buffer which are stable for hours. Isolated sugar-cane chloroplasts in this buffer retain the red versus yellow distinction for approximately an hour. Though the loss of color distinction in chloroplasts from Group I plants can be partially halted by fixation in glutaraldehyde, the use of appropriate buffers makes such treatment unnecessary. When N a O H is used to adjust the p H of the buffer, instead of KOI-I, the fluorescence recorded on the film shifts. The normally yellow mesophyll region turns greenish-yellow while the normally red bundlesheath becomes more orange. Thicker sections appear more immune to this effect, presumably because penetration is minimal in intact cells. This effect of Na + and K+ ions on fluorescence emission wavelengths m a y be related to the similar effects of ions on fluorescence (Homann, 1969; Murata, 1971).
B. Spectral Response o/In/rared Color Film Fig. l e illustrates the precise spectral response of infrared color film type 8443 to monochromatic light. The monochromatic source was a Bausch and Lomb 500 nm red blaze grating monochrometer using one m m slits. These data show t h a t this false color film records wavelengths between 660 and 690 nm as yellow, between 696 and 700 as gold, between 702 and 706 nm as orange, between 708 and 730 as red, and between 750 and 780 nm as violet. The film, therefore, distinguishes between infrared and far-red radiation with great sensitivity. Natural chloroplast fluorescence contains both far-red and infrared components. To determine the film's response to varying proportions of infrared and farred light we made a series of double exposures using 680-nm and 730-nm light (Fig. 1 f). The source was the monochrometer described above with 3-mm slits,
246
L. Elkin ~nd 1~. B. :Park
Fig. 1 a - - f
Chloroplast Fluorescence of C~ Plants I.
247
and the proportions are described in terms of total incident q u a n t a determined b y the methods of Saner and Park (1965). The film turns red-orange, instead of red with as little as a 1:4 ratio of far-red to infrared light. A true orange color develops when the ratio is 1:2. Therefore, when as little as 1/5 of the incident light is from the far-red, formation of a true red color is no longer possible. W h e n at least half of the incident light is from the far-red, formation of a true orange color is no longer possible. The gold color resulting from an equal mixture of far-red and infrared light is readily distinguishable from orange. A yellow color is not attained unless 2/3 or more of the light is from the far-red. Consequently, a yellow color on the film could either be indicative of exposure to far-red light, or of a predominance of far-red over infrared light. Similarly, a gold color results either from light between 696 and 700 nm, or from simultaneous exposure to approximately equal q u a n t u m fluxes of infrared and far-red light. An orange color indicates incident wavelengths betwen 701 and 704 nm, or else a predominance of at least 2/3 of infrared relative to a m a x i m u m of 1/3 far-red light quanta. A true red color indicates incident wavelengths of greater than 710 n m or at least 4/5 infrared light, and a m a x i m u m of 1/5 far-red light. Infrared Aero E k t a c h r o m e film 8,443 is therefore a unique tool for studying fluorescence of individual chloroplasts because it distinguishes between far-red and infrared components of fluorescence with great sensitivity. Although overor underexposure can affect the colors recorded by infrared color film 8,443, there remains a differential response to infrared versus far-red light. Overexposure maintains the distinctive response to far-red and infrared light by showing yellow-white versus orange color development; underexposure maintains it b y showing orange versus deep-red color development. The E a s t m a n K o d a k C o m p a n y has replaced Infrared Aero E k t a e h r o m e Film 8,443 with Ektaehrome Infrared Film. The emulsion of this new film is not as stable at --10 ~ as t h a t of the former product. I n addition, although fresh E k t a c h r o m e Infrared Film distinguishes between far-red and infrared radiation, the resulting colors are not nearly as distinctive as those recorded on Infrared Aero Ektaehrome Film. However, the colors associated with E k t a e h r o m e Infrared Color Film developed according to the E-4 process can be sharpened by changing the development procedure to the E-3 process (personal communication, S. H a m m o n d , H a m m o n d and Assoc., Photographers, Berkeley, California, USA).
Fig. 1 a--f. (a) Infrared fluorescence photomicrograph of a cross section of a leaf of Dichanthium annulatum, a Group I plant, photographed with Infrared Aero Ektaehrome Film. • 300. (b) Infrared fluorescence photomicrograph of a cross section of a leaf of Spartina ]oliosa, a Group II plant. )
Chloroplast Fluorescence of C~ Plants I. D e t e c t i o n w i t h I n f r a r e d Color Film* L y n n e Elkin** and t~oderie B. Park Botany Department and Lawrence Berkeley Laboratory, Berkeley, California 94720, USA Received 21 April; accepted 30 July 1975
Summary. The precise spectral responses of infrared color film permit a rapid qualitative detection of the in situ fluorescence properties of chloroplasts in CAplants. Fluorescing mesophyll and granal bundle-sheath chloroplasts appear yellow, indicating a predominance of far-red over infrared fluorescence. In contrast, fluorescing agranal bundle-sheath chloroplasts appear red, indicating a predominance of infrared over far-red fluorescence. There is a strong positive correlation between lamellar appression and the ratio of far-red to infrared fluorescence as recorded by infrared color film. The red versus yellow color of bundle-sheath chloroplast infrared photographs is used to separate the 18 species of C4 plants and 2 species of Ca plants examined into two basic groups. This separation is compatible with a separation based on lamellar ultrastrueture of bundle-sheath chloroplasts. Introduction Over the past 10 years a n u m b e r of monoeots in the Gramineae and Cyperaeeae as well as representatives from at least 10 dieot families were shown to possess the Ca p a t h w a y of carbon fixation. Characteristically, C4 plants have low photorespiration and high photosynthetic capacity. I n addition, the leaves of these plants generally display " K r a n z " a n a t o m y in which the vascular bundles are surrounded b y bundle-sheath cells. The bundle-sheath cells m a y be agranal and generally contain larger chloroplasts t h a n the surrounding mesophyll cells (see Laetseh, 1974 for review). Investigations on C4 plants have been hampered b y the difficulty of isolating and completely separating intact bundle-sheath from mesophyll chloroplasts (Laetseh, 1971, 1974). Two of the more recent studies involved improved methods for bundle-sheath preparation (Andersen et al., 1971 ; E d w a r d s and Black, 1971 ; Mayne et al., 1971), although it is questionable whether anyone has yet isolated class i (Spencer and Unt, 1965) bundlesheath chloroplasts. Alternatively, fluorescence microscopy, an excellent high-resolution, nondestructive technique f o r monitoring the photochemical reactions of chloroplasts in ~itu, m a y be used to s t u d y these chloroplasts. The fluorescence pattern m a y be quickly surveyed visually, or recorded and measured photographically. Although the green colors of mesophyll and bundle-sheath chloroplasts in sugar-cane leaves * This work was supported in part by NSF grants GB-25579X and GB-41720X and the Atomic Energy Commission, and a Sigma Xi Grant-in-Aid of Research. ** Submitted in partial requirement of the Ph.D. degree. Lynne Elkin was supported as an NII-I Predoctoral Fellow. Present address: Department of Biological Sciences, California State University at Hayward, Hayward, California 94542, U.S.A.
244
L. Elkin and 1%. B. Park
appear equally i n t e n s e to the eye when viewed u n d e r bright-field microscopy, the large b u n d l e - s h e a t h chloroplasts disappear when viewed b y fluorescence microscopy. These b u n d l e - s h e a t h chloroplasts have either a drastically reduced fluorescence yield, or a p r e d o m i n a n c e of fluorescence in the infrared region of the spectrum to which the h u m a n eye is r e l a t i v e l y insensitive, or a c o m b i n a t i o n of both. I n f r a r e d color film, a film sensitive from 500-900 nm, is capable of distinguishing b e t w e e n visible far-red a n d infrared r a d i a t i o n ( E a s t m a n K o d a k Publication M-28, 1972). I n this paper we describe the m a r k e d difference in fluorescence p a t t e r n s of sugar-cane b u n d l e - s h e a t h chloroplasts, as well as those of other Ca plants, with the use of infrared fluorescence photomicroscopy. We explore the correlation b e t w e e n the degree of lamellar appression i n the chloroplasts a n d the relative proportion of far-red to infrared fluorescence as recorded b y infrared color film.
Materials and Methods Mature, dark-green and healthy leaves are freshly detached from Mollugo cerviana Ser., and Mollugo verticillata L. of the Aizoaceae, Amaranthus edulis Miehx. ex Moq. and Froelichia gracilis (Nook.) Moq. of the Amaranthaeeae, Atriplex lenti]ormis (Torr.) W0~ts. and Spinacia oleracea L. of the Chenopodiaccae, Euphorbia maculata L., Euphorbia serphylli/olia Pers., and Euphorbia splendens Bojer of the Euphorbiaceae, Cenchrus setigerus Steud. 1. e., Cynodon dactylon (L.) Pers., Dichanthium anuulatum (Forsk.) Stapf, Digitaria sanguinalcs (L.) Scop., Echinochloa colona (L.) Link., Saccharum o]/icinarum L., Sorghum ca//forum Be~uv. and Spartina [oliosa Trin. of the Gramineae, and Portulaca oleracea L. of the Portulacaceae. Except for Dig#aria, growth conditions do not seem to affect the fluorescence properties of the plants significantly since the results were indistinguishable whether plants were obtained from growth chambers, greenhouses, gardens, or collected from the field. In contrast, in Digitaria, the bundle-sheath chloroplasts varied from brilliant red when grown in greenhouses to pale orange when grown in growth chambers. Fresh leaf sections between 12 and 36 ~m thick are cut with a razor blade or with the aid of a Lab Line/Hooker (~[elrose Park, Illinois) plant Microtome No. 1225 and are placed directly into the buffers listed below. Chloroplasts are isolated according to the procedure of Jensen and Bassham (1966). They are suspended in Jensen-Bassham solution A, adjusted to pH 6.1 with KOtI; Jensen-Bassham solution C, adjusted to pH 7.2 with KOH; or solution E which consists of potassium-phosphate buffer, adjusted to p g 7.2 with KOH and containing the same inorganic ions as the other two buffers (Elkin, 1973). It is frequently necessary to remove excess starch grains and cell debris with a preliminary 2 rain 200 g centrifugation. Specimens are examined with a Zeiss fluorescence photomicroscope which incorporates high-pressure mercury vapor lamp HBO 200 W/4, ~ KGi heat-absorbing filter, a BG38 redsuppressing filter, a BG12 exciter filter, a heat-reflecting filter, the achromatic-aplanatie phase-contrast fluorescence condenser (N.A. 1.4), apoehromatie objective lenses and a No. 53 barrier filter. The condenser is coated with oil and set on its bright-field position to optimize the collection of light. The intensity of the fluorescence striking the film is maximized by mounting a camera back (without a lens) on top of the microscope in a direct line with the specimen and the objective lens to eliminate light losses in the automatic camera attachment. Eastman Kodak Infrared Aero Ektachrome 8443 is the infrared color film exclusively used in this investigation. The spectral response of each film emulsion is tested by photographing leaf sections from sugar-cane (a C~ plant) to assure uniform color response before a batch of film is accepted for experimental use. Since infrared color film is especially sensitive to emulsion deterioration, the film must be obtained fresh and immediately stored at -- 10~ until actual use. The film must be processed or refrozen (in the presence of a desiccant) immediately after it is exposed. For proper development, infrared color film requires strict Kodak E-3 processing (Eastman Kodak, 1972).
Chloroplast Fluorescence of C~ Plants I.
245
Results
A. Chloroplast Fluorescence Figs. l a and b show two patterns of color distribution of chloroplast fluorescence as recorded by infrared color film. All higher plants examined in this study display one or the other of these patterns which we have called Groups I and II. Group I plants are typified by the pattern shown in Fig. 1 a, an infrared color photograph of fluorescing chloroplasts in a leaf cross section of Dichanthium annulatum. The bundle -sheath chloroplasts appear red and the mesophyll chloroplasts appear yellow. Group I I plants are typified by the pattern shown in Fig. l b, an infrared color photograph of fluorescing chloroplasts in a cross section of a leaf of Spartina/oliosa. The photograph contains only yellow chloroplasts. Upon careful isolation, chloroplasts from both groups maintain their color distinction, as shown in Figs. 1 c and d. The red and yellow chloroplasts in Fig. 1 c were isolated from sugar-cane, a Group I plant, while the yellow chloroplasts in Fig. 1 d were isolated from spinach, a Group I I plant. The fluorescence properties of damaged chloroplasts are different from those of intact ones. Group I mesophyll and bundle-sheath chloroplasts suspended in H~O rupture osmotically and lose the clarity of the photographic red versus yellow distinction. Similarly, the color distinctions of very thin sections suspended in H20 deteriorate within minutes, in contrast to thin sections immersed in Jensen-Bassham buffer which are stable for hours. Isolated sugar-cane chloroplasts in this buffer retain the red versus yellow distinction for approximately an hour. Though the loss of color distinction in chloroplasts from Group I plants can be partially halted by fixation in glutaraldehyde, the use of appropriate buffers makes such treatment unnecessary. When N a O H is used to adjust the p H of the buffer, instead of KOI-I, the fluorescence recorded on the film shifts. The normally yellow mesophyll region turns greenish-yellow while the normally red bundlesheath becomes more orange. Thicker sections appear more immune to this effect, presumably because penetration is minimal in intact cells. This effect of Na + and K+ ions on fluorescence emission wavelengths m a y be related to the similar effects of ions on fluorescence (Homann, 1969; Murata, 1971).
B. Spectral Response o/In/rared Color Film Fig. l e illustrates the precise spectral response of infrared color film type 8443 to monochromatic light. The monochromatic source was a Bausch and Lomb 500 nm red blaze grating monochrometer using one m m slits. These data show t h a t this false color film records wavelengths between 660 and 690 nm as yellow, between 696 and 700 as gold, between 702 and 706 nm as orange, between 708 and 730 as red, and between 750 and 780 nm as violet. The film, therefore, distinguishes between infrared and far-red radiation with great sensitivity. Natural chloroplast fluorescence contains both far-red and infrared components. To determine the film's response to varying proportions of infrared and farred light we made a series of double exposures using 680-nm and 730-nm light (Fig. 1 f). The source was the monochrometer described above with 3-mm slits,
246
L. Elkin ~nd 1~. B. :Park
Fig. 1 a - - f
Chloroplast Fluorescence of C~ Plants I.
247
and the proportions are described in terms of total incident q u a n t a determined b y the methods of Saner and Park (1965). The film turns red-orange, instead of red with as little as a 1:4 ratio of far-red to infrared light. A true orange color develops when the ratio is 1:2. Therefore, when as little as 1/5 of the incident light is from the far-red, formation of a true red color is no longer possible. W h e n at least half of the incident light is from the far-red, formation of a true orange color is no longer possible. The gold color resulting from an equal mixture of far-red and infrared light is readily distinguishable from orange. A yellow color is not attained unless 2/3 or more of the light is from the far-red. Consequently, a yellow color on the film could either be indicative of exposure to far-red light, or of a predominance of far-red over infrared light. Similarly, a gold color results either from light between 696 and 700 nm, or from simultaneous exposure to approximately equal q u a n t u m fluxes of infrared and far-red light. An orange color indicates incident wavelengths betwen 701 and 704 nm, or else a predominance of at least 2/3 of infrared relative to a m a x i m u m of 1/3 far-red light quanta. A true red color indicates incident wavelengths of greater than 710 n m or at least 4/5 infrared light, and a m a x i m u m of 1/5 far-red light. Infrared Aero E k t a c h r o m e film 8,443 is therefore a unique tool for studying fluorescence of individual chloroplasts because it distinguishes between far-red and infrared components of fluorescence with great sensitivity. Although overor underexposure can affect the colors recorded by infrared color film 8,443, there remains a differential response to infrared versus far-red light. Overexposure maintains the distinctive response to far-red and infrared light by showing yellow-white versus orange color development; underexposure maintains it b y showing orange versus deep-red color development. The E a s t m a n K o d a k C o m p a n y has replaced Infrared Aero E k t a e h r o m e Film 8,443 with Ektaehrome Infrared Film. The emulsion of this new film is not as stable at --10 ~ as t h a t of the former product. I n addition, although fresh E k t a c h r o m e Infrared Film distinguishes between far-red and infrared radiation, the resulting colors are not nearly as distinctive as those recorded on Infrared Aero Ektaehrome Film. However, the colors associated with E k t a e h r o m e Infrared Color Film developed according to the E-4 process can be sharpened by changing the development procedure to the E-3 process (personal communication, S. H a m m o n d , H a m m o n d and Assoc., Photographers, Berkeley, California, USA).
Fig. 1 a--f. (a) Infrared fluorescence photomicrograph of a cross section of a leaf of Dichanthium annulatum, a Group I plant, photographed with Infrared Aero Ektaehrome Film. • 300. (b) Infrared fluorescence photomicrograph of a cross section of a leaf of Spartina ]oliosa, a Group II plant. )
