When fresh whole leaves of six different species of forage legumes were suspended in an artificial rumen medium and inoculated with rumen bacteria, bacterial adhesion and proliferation were noted at the stomata, and penetration of the stomata by these bacteria was documented by electron microscopy. The invading bacteria adhered to surfaces within the intercellular space of the leaf and produced very extensive exopolysaccharide-enclosed microcolonies. After some of the legume leaf cell walls were disorganized and ruptured by bacterial digestion, these cells (notably, parenchyma and epidermal cells) were invaded by bacteria, with subsequent formation of intracellular microcolonies. However, other cells were neither ruptured nor colonized (notably, stomata guard cells and vascular tissue). At all stages of the digestion of intact legume leaves, the rumen bacteria grew in microcolonies composed of cells of single or mixed morphological types, and a particular ecological niche was often completely and consistently occupied by a very large microcolony of cells of single or mixed morphological types.The bacterial digestion of cut grass leaves and of cell walls prepared from various grasses has been documented by several groups (1,2,9,13,14). Brazle and Harbers (3) studied the digestion of air-dried alfalfa hay by scanning electron microscopy, but no detailed studies of the sequence of events in the bacterial digestion of fresh whole legume leaves have been reported.Various bacteria have been shown to adhere to plant cell walls to produce "pits" by their cellulolytic activity (1, 2, 5). The cell walls of different plant tissues have shown sharp differences in the extent to which they are colonized and digested by bacteria (1, 2, 5), and workers in the United Kingdom (9,13,14) and in the southern United States (1, 2) have found that morphologically different cellulolytic bacteria predominate in the digestion of plant materials in their particular geographic regions. Studies of the adhesion of groups of rumen bacteria to their polymeric substrates have shown that amylase producers adhere to starch (11) but not to cellulose and that cellulose decomposers adhere to cellulose (17, 18) but not to starch. Thus, we expect that plant material will be heavily colonized by various types of bacteria soon after it is introduced into the rumen; therefore, it is not surprising that Forsberg and Lam (10) found 75% of the adenosine triphosphate of the bacteria in the rumen contents to be associated with food particles.Because transmission electron microscopy of ruthenium red-stained sections allows both bacteria and their exopolysaccharide products to be seen throughout leaf tissues during digestion, we incubated intact legume leaves with rumen bacteria and examined them at intervals using this method. We have shown (7) that the rumen bacterial population is made up of three distinct subpopulations-the rumen fluid bacteria, the food particle-associated bacteria, and the bacteria adherent to the rumen epithelium. To obtain a good represen...

The effect of fatty acids on Escherichia coli K 1 2 was dependent on tbe source of the inoculum, the growth phase and the washing of the bacteria. The effects of saturated fatty acids from C4 to C16 and oleic acid at two concentrations (0.1 and 0.4 %, wlv) were determined on E. coli KI2/154 growing exponentially in five different culture media. Depending on the media, 0.1 % fatty acids increased the doubling times of the cultures by up to 96 %. Fatty acids of medium chain length (C6 to C I I ) at 0.4 % produced a decrease in cell concentration, nonanoic and decanoic acids being the most effective. A correlation was found between the decrease in cell concentration and the loss of viability of the cultures after addition of 0.4 % decanoic acid, with stationary-phase bacteria being affected more than those from exponential-phase cultures. Experiments carried out with E. coli B and c gave results similar to those obtained with E. coli KI2/154.

SU MMARYThe effects of acetic, propionic, butyric and lactic acids (short-chain acids), and of glycerol, on the in vitro dry matter digestibility (IVDMD) of oat hay, and on cellulolysis by a protein extract of rumen fluid were studied. The objective was to gain information on the impact of increasing the concentrations of those compounds in the medium on degradative activities carried out by rumen microorganisms. Short-chain acids were assayed as sodium salts and, for all compounds, the concentrations were 50, 100, 200 and 300 mM. The volatile fatty acids (VFA: acetic, propionic and butyric acids) were tested separately or in two mixtures A and B that contained different proportions of each VFA. The IVDMD was assayed according to a modified Tilley & Terry (1963) technique while cellulolysis was assessed by the hydrolysis of carboxymethylcellulose (CMC) in 1-h incubations at 39 xC. Parallel incubations with NaCl were utilized as controls for ionic strength and osmolarity changes in the incubation medium. Increases of 100-300 mM of short-chain acids decreased IVDMD between 7 and 39 %. The inhibitions produced by increases of 200 and 300 mM of acetic acid were smaller than those caused by the same concentrations of either butyric or lactic acids. Increases of 100, 200 and 300 mM of glycerol reduced by 8-15 % the IVDMD, and the inhibitions produced by increments of 200 and 300 mM of this compound were smaller than the ones elicited by the shortchain acids, with the exception of 200 mM of acetic acid that did not differ. The IVDMD was inhibited by NaCl only when concentrations were increased by 200 (12 %) and 300 mM (26 %). The degradation of CMC was reduced by 7-20 % by 200 and 300 mM increases of the short-chain acids or glycerol, the exception being lactic acid that had no effect. One hundred mM increases of acetic and butyric acids inhibited the CMC degradation by 7 and 9 %, respectively, whereas only butyric acid was inhibitory (7 %) at 50 mM. Greater concentration increases of both VFA mixtures A and B than of the individual VFA were necessary to inhibit the hydrolysis of CMC. Cellulolytic activity was decreased 6 and 9 % by increases of 200 and 300 mM of NaCl, respectively, and these drops of activity were smaller than those obtained with increases in any of the other compounds. Since osmolarity and/or ionic strength changes in the medium cannot completely account for the observed inhibitions of IVDMD and cellulolysis, it is suggested that glycerol and the anions of short-chain acids produce changes in the reaction media polarity that contribute to the inhibitory effects observed. Alterations in the media could also bring about conformational changes in the degradative enzymes leading to reduced rates of reaction and subsequent decreases in IVDMD and cellulolysis. Since most of the effects were observed with concentration increases that were larger than the physiological concentrations in the rumen, it is suggested that the compounds tested have little impact in vivo on the ruminal degradative activities that w...