Glasshouse experiments with Ricinus communis showed that the presence/absence of a VA mycorrhizal fungus (Glomus clarum) changed the δ15N value of the host by as much as 2‰ when the plants were given urea (released as NH4+) as their only N‐source. This small change in Δ15N would create a large error in calculating sources of plant N. In particular, these results throw into doubt any models of N‐cycling which assume that soil N can be treated as a single source. The correct N‐source value for VAM‐infected NH4− ‐using plants may be the δ15N of soil NH4++ 2‰. Treatment effects were also found in the distribution of δ15N and % N among plant organs. Plants with VAM had a lower N:P atom ratio and were larger in total biomass. Carbon discrimination (δ13C) was greater in the VA‐infected plants. The measured effects of VAM infection suggest that for some plants the fungus may be the primary site of N assimilation. A parallel experiment with Eucalyptus globulus and the ectomycorrhizal fungus Hydnangium carneum resulted in no significant differences in any of the variables measured for this host‐fungus pair when the sole N‐sources were inorganic (NO3− and NH4+ released from urea). Ectomycorrhizal fungi are diverse in their physiological behaviour, and these data should not be taken as being representative of the whole group. More work is required with other types of mycorrhiza and more complex sources of N. Future work will include a water balance to partition the effects of water use and nutrient supply in determining δ13C. An on‐line combustion‐ANCA‐MS method is described for fully automated measurement of natural abundance levels of 15/14N and 13/12C for plant materials. This method achieves the required precision while dramatically increasing sample throughout.

SUMMARYThe growth effects of three mycorrhizal Endogone endophytes on tobacco, tomato and maize were examined. The extent of the stimulus obtained depended on nutrient conditions and the subsequent level of infection developed in the root system. Marked stimulation occurred with low phosphate availability and high root infection.

Algal-iysing bacteria are present in five lochs, two reservoirs, one sewage works and five alkaline soils whieh were examined. Nine out of thirty-six isolates have been studied in detail and all are gram-negative non-fruiting myxobaeteria. All are aerophilic, require a neutral or alkaline pH (optimum 7-0-9 0) for good growth, have temperature optima of 28-37 "C, require contact with the host for lysis to oecur, and are so similar in morphology and host range that they all appear to be strains of a single genus. They lyse all bloom-forming Cyanophyeeae on whieh they have been tested, including species of Anabaena, Aphanizomenon, Gloeotrichia, Microcystis and Oscillatoria. Their abundance is usually correlated direetly with the abundance of Cyanophyeeae in eutrophic waters and it is probable that in most ecosystems the algae and the bacteria co-exist, although if the equilibrium changes markedly as a result of change in environmental conditions situations may arise where the bacteria may play an important role in the lysis of algal blooms. The available evidence suggests, however, that this is likely to be a rare occurrence in natural ecosystems.Ecological studies on algal-lysing bacteria in fresh waters 579 Materials and MethodsHabitats studied Eight fresh waters in Scotland (Balgavies Loch, Long Loch, Loch of Lindores, Monikie Reservoir and Thriepley Loch), England (Esthwaite Water and Loweswater) and Wales (Eglwys Nunydd Reservoir) were surveyed for the presence of iytic bacteria. These habitats are detailed in Table 1. The four Scottish lochs are mesotrophic or eutrophic, are dominated by species of Anabaena, Aphanizomenon, Microcystis and Asterionella and are used for angling and recreational sports. The waterworks at Monikie belong to the East of Scotland Water Board and the slow sand filters often develop a succession of algae during the summer in which some, or all of Gloeotrichia echintdata, Microcystis aeruginosa, Aphanizomenon flos-aquae and Anabaena circinalis predominate. Eglwys Nunydd Reservoir, in Wales, provides water for the British Steel Corporation strip mills at Port Talbot. Its flora is usually dominated in summer by Microcystis. Esthwaite Water and Loweswater are eutrophic waters in the English Lake District, and were sampled only once (6 October, 1971) for the presence of algal-lysing bacteria. During this sampling period Microcystis was present in Esthwaite Water and Gomphosphaeria and Anabaena were present in Loweswater. The sewage works at Forfar, Angus (Nat. Grid. Ref. NO 443502), which is of the trickling filter type, and various soils in the vicinity of Dundee were also sampled.

Summary Four bacterial isolates have been obtained which cause lysis of laboratory cultures and natural populations of blue‐green algae. Over forty algal strains belonging to all orders of the blue‐green algae and including bloom‐forming species are susceptible. The bacteria also lyse certain gram‐positive and gram‐negative bacteria. The organisms are aerobic, gram‐negative rods, aflagellate, motile and pigmented with an absorption maximum of the pigments in methanol occurring at 435 nm and with shoulders at 460 nm and 415 nm. The guanine and cytosine in the DNA is within the range of 65–69± 1 M%. The organisms are identified tentatively as members of the Myxobacterales. Lysis of vegetative cells of the algae depends on the presence of growing bacteria, bacteria‐free filtrates alone being insufficient to cause lysis. Heterocysts are not lysed. The bacterial plaques resemble viral plaques in appearance. The lytic agents rapidly inhibit metabolic activity of the susceptible algae and may cause cell lysis within 2–10 hours. The isolates can be propagated readily in the absence of the host and are probably widely distributed in freshwater habitats.

SUMMARYVarying amounts of soluble phosphate were applied over different periods during the growth of mycorrhizal maize. The amount of the endophyte produced within the root system and the growth of the host were related inversely. Repeated small doses of soluble phosphate over long periods depressed the production of the fungus more than when given over shorter periods. Large single applications of soluble phosphate applied early in the growth of the host also reduced the amount of the endophyte more than when given later. The role of mycorrhiza in relation to phosphate nutrition is discussed.