Brachypodium Sylvaticum Research Articles

Contrasting patterns of photosynthetic acclimation to the light environment are dependent on the differential expression of the responses to altered irradiance and spectral quality

Chl, chlorophyll Chl a/b, ratio of chlorophyll a to chlorophyll b Cyt f, cytochrome f FR, far‐red light LFR, low irradiance, far‐red enriched growth light LHCII, light harvesting complex associated with PSII LW, low irradiance, white growth light MW, moderate irradiance, white growth light PAR, photosynthetically active radiation Pmax, light and CO2 saturated photosynthetic rate PSI, photosystem I PSII, photosystem II Four plant species (Chamerion angustifolium, Digitalis purpurea, Brachypodium sylvaticum and Plantago lanceolata) which have previously been shown to demonstrate contrasting photosynthetic acclimatory responses to the light environment (Murchie & Horton 1997, Plant, Cell and Environment 20, pp. 438–448) were analysed at a biochemical level. Plants were grown under low irradiance with a shade‐type spectrum (LFR: 50μmol quanta m–2 s–1), moderately high white light (MW: 300μmol quanta m–2 s–1) and low irradiance white light (LW: 50μmol quanta m–2 s–1). The effects of light quality upon chlorophyll content and photosynthetic capacity were found to be species‐dependent. A far‐red dependent reduction in chlorophyll was found in three species, and an irradiance‐dependent reduction was found in B. sylvaticum, which showed the greatest alteration in the xanthophyll cycle pool size of all species tested under these conditions. Chlorophyll a/b ratios were sensitive to both light quality and quantity in C. angustifolium and D. purpurea, being highest in MW, lowest in LFR, and intermediate in LW, whilst the other species showed no response. Ratios of photosystem II to photosystem I (PSII and PSI) demonstrated a strong irradiance‐associated increase in all species except B. sylvaticum, whereas an increase in PSII/PSI in LFR compared to LW conditions was present in all species. A change in chlorophyll a/b was not always associated with a change in PSII/PSI, suggesting that the level of LHCII associated with each PSII varied in some species. Cytochrome f content showed an irradiance‐dependent effect only, indicating a relationship with the capacity of electron transport. It is concluded that differing strategies of acclimation to the light environment demonstrated by these species results from differing strengths of expression of a series of independently regulated changes in the levels of photosynthetic components.

Observations on patterns of granular fertiliser deposition beside hedges and its likely effects on the botanical composition of field margins

SummaryPatterns of granular fertiliser deposition differed markedly between a pneumatic boom applicator and a spinning disk machine, both of which were tractor‐mounted. The pneumatic applicator gave relatively even distribution across the boom width. A spinning disk gave more variable deposition and significant amounts were spread further than the expected 12 m overlapping pattern. When operating with a deflector plate on the spinning disk, significant amounts of fertiliser were also spread further than the expected 6 m. However, when operating next to a hedge, there was evidence of fertiliser concentration at the base of the hedge and prevention of granules passing through to adjacent habitat. A peak concentration of up to 150 kg ha‐1 of fertiliser was deposited at the hedge/field edge interface.In a tray experiment on seedling competition and establishment, there was no evidence of nitrogen fertiliser effects, at rates found in the field, on early plant establishment or species diversity. However, in a competition experiment with established plants of four grass species grown in pots, the nitrophilous species Bromus sterilis was able to increase growth at increasing nitrogen level, at the expense of slow‐growing Brachypodium sylvaticum and Anthoxanthum odoratum. Whilst hedges may buffer fertiliser contamination of habitats adjacent to agricultural fields, deposition of concentrated nitrogen fertiliser beside the hedge is likely to encourage the growth of nitrophilous plant species. In the short term, this may not affect botanical composition or diversity, but to reduce the likely long‐term adverse effects of fertiliser misplacement, we recommend that farmers are encouraged to use pneumatic fertiliser applicators and to introduce vegetation buffer strips at field edges.

Ecophysiology of five Mediterranean perennial grasses: II) Effects of shade on gas exchange

ABSTRACT Five perennial grasses characteristic of different Mediterranean environments (Ampelodesmos mauritanicus (Poiret) Dur. et Sch., Brachypodium ramosum (L.) R. et S., Festuca drymeia M. et K., Brachypodium sylvaticum (Hudson) Beauv., Festuca heterophylla Lam.) were pot cultivated at two different levels of light intensity. Gas exchanges were studied by a Li-Cor 6400 Portable Photosynthesis System. Light response curves revealed a greater photosynthetic potential for the species of open habitats, though maximum carbon uptake was much lower among shaded replicates. On the contrary, the species from shaded habitats exhibited lower net photosynthesis at saturation but small differences between plants grown in light or shade. The five species showed relevant differences with respect to transpiration and water use efficiency. Results provide useful information on the ecophysiology of these species in relation to their different field distribution.

Host plant chemistry and preferences in egg-laying Trioza apicalis (Homoptera, Psylloidea)

Sixtee n belonging to the family Apiaceae were examined for their acceptance as host plants by Trioza apicalis in non-choice tests. The number of eggs laid per day varied from 18 on Daucus carota subsp, sativus to zero on Aegopodium podagraria. In a comparative test, the psyllids landed significantly faster and started egg-laying earlier on carrots (D. carota subsp, sativus) than on dill (Anethum graveolens). In a study of shelter plants for T. apicalis, the psyllids survived on spruce, Picea abies, for more than 30weeks, but they died on grass, Phleum pratense and Brachypodium sylvaticum, within 2 weeks. The length of the day is one of the factors regulating hibernation. At 20°C, light regimes shorter than 17 h induced hibernation, while longer light periods led to reproduction. Both the shelter plants and the summer hosts, including the Apiaceae species, released large amounts of monoterpene hydrocarbons, with great differences among the species. The most preferred host plants (carrots) were found to contain (+)- and (-)-c~-pinene and (+)-sabinene, while large amounts ( > 20%) of either (+) - or ( - ) -limonene were released by the species of low preference. ( - ).-c~- Pinene, ( - ) -J3-pinene, and (-) -limonene were the main components of the spruce odour. A multivariate data analysis of the chemical results showed clearly the significant similarities of the summer host and shelter plants (Daucus carota, Picea abies, and Juniperus communis). © 1997 Elsevier Science Ltd

Ecophysiology of five Mediterranean perennial grasses: I) Effects of shade, water stress and defoliation on growth and allocation

ABSTRACT Five perennial grasses from different Mediterranean environments (Ampelodesmos mauritanicus (Poiret) Dur et Sch., Brachypodium ramosutn (L.) R. et S., Festuca drymeia M. et K., Brachypodium sylvaticum (Hudson) Beauv., Festuca heterophylla Lam.) were pot cultivated and subjected to shading, water stress and defoliation in a factorial experiment. Growth, root/shoot ratio and reproductive allocation were measured during 7 months. Drought conditions determined a growth reduction in all species, though A. mauritanicus and F. heterophylla showed a relatively higher tolerance. Shade effects were evident under high watering regime, but they were not significant for droughted plants. Highest compensatory growth after defoliation was shown by B. ramosum under full light and by B. sylvaticum under shade. The effects of the experimental treatments are discussed in this paper in relation to the different ecological behaviour of the five grasses.

EVIDENCE FOR GENETIC DIFFERENTIATION BETWEEN CHOKE-INDUCING AND ASYMPTOMATIC STRAINS OF THE EPICHLOË GRASS ENDOPHYTE FROM BRACHYPODIUM SYLVATICUM.

Life cycle and breeding system variation in Epichloë grass endophytes (choke disease) is tightly linked to the degree of stroma formation. It is not known whether this variation results from differences in host resistance, fungal virulence, or environmental conditions. We found genetic differentiation between 173 asymptomatic (NS) and 93 stromata-forming (S) Epichloë strains isolated from one grass species, Brachypodium sylvaticum, based on 13 presumed allozyme loci, of which six were variable. The fungal strains originated from 10 sites in Switzerland, three sites of which were represented by both NS and S subpopulations. In total, 19 allozyme genotypes, that were nonrandomly distributed among S and NS were detected. Genetic variation measured as GST between S and NS strains isolated from the same site ranged from 0.73 to 0.98. Clonality, measured as linkage disequilibrium at one site, was significant in the NS subpopulation (P ≪ 0.001), but not in the S subpopulation (P = 0.21), implying asexual reproduction by NS strains as well as successful horizontal transmission of S strains. Since all seeds are usually infected vegetatively, horizontal transmission implies the occurrence of multiple host infections. Altogether, these results provide indirect evidence that NS and S strains do not belong to one panmictic population and that differentiation patterns of stroma formation found in nature are due to genetic differences among fungi in associations with their host plants. We discuss the direction of evolution of disease expression in this system. The distribution of genetic variability suggests that the asymptomatic strains were derived from stromata-forming populations.

Evidence for Genetic Differentiation Between Choke-Inducing and Asymptomtic Strains of the Epichloe Grass Endophyte from Brachypodium sylvaticum

Evidence for Genetic Differentiation Between Choke-Inducing and Asymptomtic Strains of the Epichloe Grass Endophyte from Brachypodium sylvaticum

Experimental evidence for the relative sensitivity of deciduous forest plants to high soil acidity

Thirteen species of deciduous forest vascular plants were established from seeds: Brachypodium sylvaticum, Carex sylvatica, Circaea lutetiana, Deschampsia flexuosa, Galium odoratum, Lamium galeobdolon, Lathyrus vernus, Luzula pilosa, Milium effusum, Poa nemoralis, Sanicula europaea, Stellaria holostea and Stellaria nemorum. An acid beech forest soil was variously treated with carbonates (CaCO 3 + MgCO 3 or SrCO 3) to obtain a soil solution pH of 3.3–5.2. The experimental results were compared with the field distribution of the same species in 230 south Swedish deciduous forests with regard to soil pH. In general, the species reacted to increasing soil pH in accordance with their field distribution. Addition of SrCO 3 was almost equally effective as CaCO 3 + MgCO 3 (both at pH 4.2 in the soil solution) in promoting establishment and growth, whereas most species failed in the untreated soil (pH 3.3). There were no indications that the Ca or Mg concentrations of the soil solution were below minimum or even suboptimal for plant growth in the untreated soil. Within the pH range (pH ⩽ 3.5–4.3) which proved critical to establishment and growth in most of the species, the reactive Al 3+ concentration of the soil solution did not vary significantly. It is suggested that the high acidity of the rhizosphere solution, possibly aggravated by Al 3+, obstructed the uptake or retention of mineral nutrients by the roots in the untreated acid beech forest soil.

Performance of Two Contrasting Morphs of Brachypodium Sylvaticum Transplanted into Shaded and Unshaded Sites

SUMMARY by the presence or absence of spreading macro-hairs at the leaf margins, termed hairy-edge (HE) and non-hairy-edge (NHE), are described. The two morphs differ in their field distribution in respect of vegetational shade; the NHE morph is confined to shaded sites, whereas the HE morph is found in both open and shaded habitats but is generally the less-frequent morph in shaded conditions. (2) Plants of the two morphs were transplanted into a deciduous woodland and into an adjacent unshaded grassland and their survival and growth monitored over a 27-month period. (3) At the woodland site 77% of the NHE transplants survived to 27 months compared with 45% of the HE plants. In contrast, at the unshaded site, 90% of the HE plants survived to 27 months compared with 60% of the NHE plants. (4) There were large differences in a number of measures of growth between the two morphs at each site. The NHE morph grew better at the shaded site whilst the HE morph grew better at the unshaded site. The NHE morph allocated a greater proportion of the above-ground dry matter to reproductive tillers than the HE morph at both sites. (5) The fitness of the 'alien morph' at each site (HE in the shaded site; NHE in the open site) was calculated relative to that of the 'native morph'. The relative fitness of the alien morph declined with time for survival and tiller number and was also low for characters measured at the final harvest. The values indicate that strong selection pressures operate against the HE morph in shaded habitats and against the NHE morph in unshaded habitats.

Johnsongrass Chlorotic Stripe Mosaic and its Associated Viruslike Particles in Iran1

AbstractA mosaic disease of Johnsongrass characterized by the presence of short and long chlorotic stripes in leaves and stunting of plants was found in two locations about 300 km apart in the Fars Province of Iran. The disease occurred in patches or along the irrigation ditches. The disease agent could not be transmitted mechanically or by certain insects and mites. About 0.5% of the seedlings, grown in soil from the root zone of diseased plants, developed the disease after 1–8 months. Electron microscopy, of leaf‐dip preparations showed presence of isometric viruslike particles (VLPs) of about 35 nm diameter in unusually high concentrations in diseased plants but not in healthy plants. VLPs from diseased leaves were purified by low‐pH clarification followed by differential and densitygradient centrifugation. Purified VLP preparations showed UV absorption spectrum typical of nucleoproteins. A260/A280 was 1.48. High titered antisera obtained by injecting rabbits with purified VLPs reacted with sap from diseased plants but not healthy plants. Purified VLP preparations and sap from diseased plants did not react with antisera to Bermuda grass etched line, Brachypodium sylvaticum, brome mosaic, maize chlorotic dwarf, maize rayado fino, maize white line mosaic and tobacco necrosis viruses.

Evidence from ascospores for host restriction in Claviceps purpurea

On the assumption that there may be significant differences in the length of ascospores of Claviceps purpurea from different hosts, and that such differences are primarily under genetic control, a comparative study was made of 85 samples of ascospores obtained from ergots that had been collected from 28 species of British grasses belonging to 24 genera and 10 tribes. Because it is shown that different samples of ascospores from the same hosts could reasonably represent parts of the same population, it is possible to separate populations of ascospores obtained from different hosts into groups by using Duncan's multiple range test. On the basis of the mean length of their ascospores, the population means for 17 different grasses, covering 16 different genera, may be divided at a probability level of 005% into three groups. One group (with ascospores averaging > 100 μm long) includes the strains of C. purpurea infecting Triticum aestivum, Glyceria fluitans and Brachypodium sylvaticum , a second group (with ascospores averaging 79–91 μm long) contains the strains infecting all the other grasses included in the test except Deschampsia caespitosa , which is the only member of a third group (with ascospores averaging C. purpurea infecting the four abovementioned grasses are host-restricted. This conclusion is discussed in relation to the results of cross-inoculation experiments, and alkaloid analyses of ergot sclerotia, carried out by other workers.

Estimation of the light distribution between photosystems I and II in intact wheat leaves by fluorescence and photoacoustic measurements

Usisng intact leaves, the extent of the decrease in chlorophyll a fluorescenece caused by the addition of continuous 710 nm light superimposed on modulated (20 Hz) 550 nm light was used to determine the distribution of this absorbed light between photosystems I (α) and II (β). The Fo and Fm levels, which defined the total variable fluorescenece, were taken as equal to those obtained with excess 710 nm light and with saturating blue-green light, respectively.An analogous procedure was used with a photoacoustic detector, saturating white light defining a base line for oxygen yield, the levels with an without 710 nm light being used to define β and α respectively.The two methods gave similar values for the distribution of light between the two photosystems for the experimental conditions used, β averaging 0.55 for a range of Triticum genotypes and Brachypodium sylvaticum grown in high or low light.

Some properties of a virus‐like agent found in Brachypodium sylvaticum in the United Kingdom

A spherical virus–like agent ($$32 nm in diameter) was isolated from a plant of Brachypodium sylvaticum (Huds.) Beauv. growing in a botanic garden in England and showing yellow streaks in the leaves. The agent was readily purified and sedimented as a single component in sucrose density rate gradients. The particles had a sedimentation coefficient at infinite dilution of $$122S and a buoyant density of 1.35 g/cm3 in CsCl and 1·33 in CS2SO4. The particles were stable at acid pH but above pH 7·0 in the presence of EDTA dissociated. A protein having a major polypeptide with a molecular weight of $$3·76 × 104 and a species of single stranded RN A with a MW of 1·67 × 106 were detected in the particles. The agent was not transmitted by manual inoculation, by the insects Myzus persicae Sulzer, Rhopalosiphum padi L. or Nephotettix virescens Distant, through soil by leakage from roots or by seed. The particles had physicochemical properties in common with tombus– and sobemoviruses but were not serologically related to 10 members of these groups or to 57 other small spherical RNA plant viruses.

Isolation and partial characterization of a lectin from a false brome grass (Brachypodium sylvaticum).

A lectin has been isolated from embryos of a false brome grass species (Brachypodium sylvaticum) by affinity chromatography on immobilized N-acetylglucosamine. It is a dimeric protein of two identical subunits of mol.wt. 18 000. Although it resembles cereal lectins with respect to its biochemical and physicochemical properties, it differs structurally in several aspects from wheat-germ-agglutinin-like lectins.

Puccinia Velutina: A New Caricicolous Rust and A Perfect State of Aecidium Elaeagni

Reciprocal inoculations with basidiospores and aeciospores proved that the telial host of Aecidium elaeagni on Elaeagnus species is Carex lenta. Comparative morphology with other caricicolous rusts shows that the rust fungus is new. We name it Puccinia velutina Four Aecidium species parasitic on Elaeagnus were reported in Japan (Ito, 1950; Hiratsuka, 1960). We found Aecidium elaeagni Diet. on Elaeagnus glabra Thunb. at Mt. Tsukuba, Ibaraki, Japan. It was described on E. pungens Thunb. (Dietel, 1898), and is widely distributed in Japan, except Hokkaido, and in Taiwan (Ito, 1950; Hiratsuka, 1960). This rust has been considered as the spermogonial and aecial stage of Puccinia coronata Corda. We therefore collected teliospores of P. coronata on Calamagrostis arundinacea (L.) Roth var. brachytricha (Steud.) Hack. and Brachypodium sylvaticum (Huds.) Beauv. at Mt. Tsukuba, and inoculated Elaeagnus species with basidiospores from teliospores of P. coronata from these collections. However, results were negative. In 1979, we found the telial stage of Puccinia sp. on Carex lenta D. Don near Aecidium-infected E. glabra, and suspected a genetic relation between the rusts on E. glabra and C. lenta. We report here the results of reciprocal inoculations and morphological observations of the rusts, and discuss their taxonomy. INOCULATION EXPERIMENTS

The annual diet of the Roe deer (Capreolus capreolus) in the New Forest, Hampshire, as determined by rumen content analysis

By analysing rumen content samples from 105 dead specimens, the annual diet of the Roe deer (Capreolus capreolus) in the New Forest, Hampshire, was determined over the period from November 1970 to March 1973. Bramble and rose formed the bulk of the diet throughout the year, comprising the largest food fraction in all months except January and April. In January to March, foliage from felled conifers, Calluna, and ivy were major foods, plus lesser amounts of grasses, herbs and fungi. From May until early winter, herbs and grasses were important. New, tender, spring and summer growth from broadleaved deciduous trees, bilberry and holly was eaten in appreciable quantities. Acorns were a characteristic autumnal food. Abundant potential foods which Roe deer seldom selected included rushes, sedges, mosses, bark and dead hardwood leaves. Widespread species that Roe rarely consumed were Corsican pine, Erica tetralix, E. cinerea, gorses, bog myrtle and the grasses Agrostis setacea, Brachypodium sylvaticum, Deschampsia caespitosa and Molinia caerulea.

THE INABILITY OF BRACHYPODIUM SYLVATICUM AND OTHER SPECIES TO UTILIZE APATITE OR ORGANICALLY BOUND PHOSPHATE IN CALCAREOUS SOILS

SummarySeedlings of Brachypodium sylvaticum (Huds.) Beauv., Festuca gigantea (L.) Vill., Lolium perenne L. and Urtica dioica L. were grown on freshly collected soil from the upper horizons of an uncultivated rendzina and a brown forest‐soil. All species responded to addition of a soluble phosphate on both soils but, while there was no response to the addition of apatite on the rendzina, there was as large a response to apatite as to soluble phosphate on the brown forest‐soil. Brachypodium sylvaticum and Urtica dioica were also grown in sand‐culture without phosphate, with a soluble phosphate, and with humus which is rich in organic phosphate. Both species responded to soluble phosphate but showed no response to the presence of humus. The experiments imply that even Brachypodium sylvaticum, which may occur naturally on calcareous soils containing only traces of soluble phosphate, cannot obtain phosphate from either apatite or humus when the pH is above 7.0.

On Morphological Convergence between <i>Brachypodium sylvaticum</i> and <i>Agropyron yezoense</i>

The remarkable resemblance in spikes and spikelets between Brachypodium sylvaticum and Agropyron yezoense is pointed out. The comparison of chromosomes between the species of Brachypodium and those of other genera of Hordeeae suggests that this resemblance is due to convergent evolution.I wish to express my cordial thanks to Dr. J. Ohwi who kindly examined the manuscript of the present paper. Thanks are also due to Dr. E. Potztal and Dr. Y. Takenaka for their kindness during the course of the present investigations.

Interspecific and intergeneric hybrids in herbage grasses xviii. Various crosses includingLolium rigidum sens.ampl withL. Temulentum andL. Loliageum withFestuga pratensis and withF. Arundinagea

1. Negative results have been recorded for the following crosses: Bromus asper ♀ x Brachypodium sylvaticum ♂, Brachypodium sylvaticum ♀ x Bromus asper ♂, Avena pratensis ♀ x Bromus erectus ♂, Bromus erectus ♀ x Festuca arundinacea ♂, Lolium loliaceum ♀ x Lepturus cylindricus ♂. 2. The tests for all these crosses were limited so that the negative results can only show that the species pairs concerned are not readily interfertile. 3. No fully establishedFb1 hybrids were obtained whenL. temulentum was pollinated by aL. rigidum s.a. plant (which would be classifiable asL. multiflorum Lam.), but a single mature plant was produced when the cross was made in the opposite direction. 4. These results are in line with those obtained whenL. temulentum andL. rigidum Gaud. were intercrossed, but differ from those given byL. temulentum intercrossed withL. italicum orL. perenne. 5. TheFb1 hybrid from the crossL. rigidum s.a.♀ xL. temulentum ♂, plant 666-bE-1, was effectively pollen-sterile but gave established progeny when used as the pistillate parent in backcrosses to either parental type. 6. From the backcross to theL. rigidum s.a. parental line, four plants were obtained, all of which could be used as either pistillate or pollen parents, although at least one of them still showed evidence of hybridity. 7. Another of the four backcross plants behaved as a normalL. rigidum s.a. plant of the parental line and had inherited the ‘brittle-stem’ character carried by this line as a simple monohybrid recessive. 8. The results obtained by intercrossingL. italicum withF. pratensis were in line with those previously reported forL. perenne withF. pratensis, but in the present case no fully matureFb1 hybrid plants were obtained. 9. Germinable seeds but no established plants resulted from the pollination ofL. rigidum s.a. byF. pratensis. 10. Caryopses were produced whenF. pratensis andL. temulentum were intercrossed in either direction, and one seed made an abortive attempt to germinate. 11. Caryopses were also produced whenF. pratensis was pollinated byL. loliaceum but none of these seeds germinated. 12. WhenL. loliaceum was pollinated byF. pratensis, seed-setting, seed germination andFb1 hybrid plant establishment were all relatively high as compared with other crosses in whichLolium species were pollinated byF. pratensis. 13. Considering the extreme morphological and other differences betweenL. loliaceum (a diploid) andF. arundinacea (a hexaploid) the positive results obtained in seed-setting, seed germination andFb1 hybrid plant establishment whenL. loliaceum was used as the pistillate parent were remarkable. 14. In the discussion it is concluded that while it may be assumed that a degree of intercompatibility between two species sufficient to produce definite caryopses is indicative of phylogenetic relationship, the measure of this relationship is not directly represented by the relative ease with which established Fp1 hybrids can be produced.

Interspecific and intergeneric hybrids in herbage grasses xvii. Further crosses involvingLolium peeenne

1. Completely negative results were obtained in limited attempts to intercrossLolium perenne withArrhenatherum avenaceum, Brachypodium sylvaticum, Bromus eretus andDactylis glomerata. 2. The results of intercrossingL. perenne withBromus asper and withLepturus cylindricus must also be regarded as negative. 3. One seedling was obtained by intercrossingL. perenne andGlyceria fluitans, but this plant produced no inflorescences and its constitution is not known. 4. EstablishedFb1 hybrid plants were obtained whenL. perenne was pollinated byFesluca gigantea, but seeds from the reciprocal cross failed to germinate. 5. The anthers of theL. perenne-F. gigantea Fp1 hybrids were non-dehiscent, and in limited tests the ovaries did not respond to pollination by either parent species. 6. Additional crosses betweenL. perenne andF. arundinacea confirmed results previously reported. WithF. arundinacea as the pistillate parent the seeds produced failed to germinate. 7. The anthers of theL. perenne-F. arundinacea Fp1 hybrids were non-dehiscent, but one plant gave established progeny when pollinated by eitherF. arundinacea or byL. perenne. 8. The single backcross plant obtained by the pollination of the Fp1 hybrid byF. arundinacea was highly sterile and gave no progeny. 9. The backcross plants obtained by pollinating theL. perenne-F. arundinacea Fp1 hybrid withL. perenne pollen were highly sterile, but one of them, the 22-chromosome plant 23-bE-7/1, gave established progeny when pollinated by eitherF. arundinacea or byL. perenne. 10. The single plant obtained by the pollination of plant 23-bE-7/1 withF. arundinacea gave no progeny. 11. Plant 23-bE-7/l was also successfully intercrossed with an artificially produced but unrelated triploid hybrid derivative, plant 56-bE-2. 12. Plants 23-bE-7/1 and 56-bE-2 and their common derivatives have been discussed elsewhere in another connexion (Jenkin, 1955f). 13. Some of the derivatives of plant 23-bE-7/1 backcrossed toL. perenne were used for further breeding, and both the cytological and pollen-fertility data given by Peto (1933) have been used in discussing the results. 14. The plants of the third generation (second backcross toL. perenne) varied considerably both in type and in vigour, although in no case was the influence ofF. arundinacea ancestry strongly evident. Some were morphologically indistinguishable fromL. perenne. 15. Some of the least ‘normal’ (onL. perenne standards) were not studied by Peto and therefore his results, 2n=14 to 2n=18, do not necessarily represent the group as a whole. 16. Plants of the third generation with14, 17 and 18 chromosomes respectively were successfully intercrossed withF. arundinacea; plants with15, 17 and 18 chromosomes were also successfully intercrossed withL. perenne, while 14- and 15-chromosome plants when intercrossed gave established progeny. 17. Plants of the fourth generation (third backcross toL. perenne) were, with a few exceptions, indistinguishable morphologically fromL. perenne. The plants of this generation examined by Peto were of three chromosome types, 2n=16 (one plant, derived from an 18-chromosome parent); 2n=15 (six plants) and 2n=14 (seventeen plants). 18. With repeated backcrossing toL. perenne there was a progressive return towards this species in both morphology and chromosome numbers. 19. There was also a return towards full fertility, but there were exceptional cases where chromosome numbers and fertility did not correspond. 20. It is concluded that natural intercrossing ofL. perenne andF. arundinacea may cause a temporary disturbance of a plant population but that it has no appreciable permanent effect unless exceptional circumstances arise.