HORTSCIENCE 26(10):1283-1284. 1991. quate for roses. However, all nutrients, ex- cept Ca and Mg, were present in far higher concentrations in the gall and in the associ- Galls Induced by Diplolepis spinosa ated leaf and stem tissues than they were in the nongalled stems. Extremely high con- Influence Distribution of Mineral centrations of Cu, Zn, Fe, and Ni were found in the stunted leaves proximal to the galls Nutrients in the Shrub Rose (Table 1). Levels of mineral nutrients in the gall leaves were higher than levels in the G. Bagatto, T.J. Zmijowskyj, and J.D. Shorthousel galls themselves. The leaf concentrations for Department of Biology, Laurentian University, Sudbury, Ont. Mn were similar for normal leaves and gall P3E 2C6, Canada leaves. The higher levels of some nutrients in Additional index words. , Cynipidae, Rosa rugosa, plant stress leaves associated with the galls than the galls are difficult to explain. We suspect that galls Abstract. Mineral nutrient levels in tissues of the domestic shrub rose (Rosa rugosa of D. spinosa and the larvae within direct Thunb.) were examined following gall induction by the cynipid wasp Diplolepis spinosa movement of nutrients from the host plant (Ashmead). Higher levels of Cu, Zn, Fe, Mn, and Ni were found in the galls, leaves through the xylem of the gall and eventually protuding from the galls, and stems to which galls are attached than were found in into the leaves; however, remobilization from normal stems and leaves, indicating that cynipid galls alter the nutritional status of the gall leaves may be prevented because of their hosts. Cynipid galls must be removed or controlled because the stress they cause modified phloem flow. Furthermore, gall will increase the vulnerability of roses to attack by other pestiferous . leaves become chlorotic and necrotic, thus, an accumulation of mineral nutrients and in- Two species of gall-inducing cynipid wasps ical galls in the maturation phase. Each of terrupted vascular flow may account for this [Diplolepis spinosa and Diplolepis radicum the stems was 30 cm long and was re- appearance: Histological studies of stem, gall, (Osten Sacken)], normally found only on the moved from an urban roadside flower bed in and associated leaf tissues could examine this wild rose (Rosa blanda Ait.), are now at- Sudbury, Ont. The normal stems were sep- possibility. tacking the domestic shrub rose in Canada arated into leaves and the top 5 cm of the The preliminary findings reported here il- and the northern United States (Shorthouse, stem. The galled stems were separated into lustrate that galls of D. spinosa alter the min- 1988). Although galls of both species alter a 5-cm sample of stem below the gall, leaf eral nutritional status of their host plants. the aesthetic appearance of shrub roses, there tissues associated with the gall, and tissues They provide further evidence that galls are have been no attempts to determine whether of the gall itself. Stem and leaf samples for or not these insects cause structural or phys- each category of tissue were pooled, oven- iological damage. Most gall-inducing insects dried, homogenized in a plant mill, and di- are thought not to cause host damage; how- gested in boiling aqua regia, and elemental ever, there are studies showing that galls are analysis was performed by flame atomic ab- physiological sinks for assimilates and sorption spectrophotometry. Gall tissues were macronutrients (Abrahamson and Weis, not pooled, and each was analyzed sepa- 1987). Here we report on how one of these rately. Details of the analytical technique are insects, the stem-galler Diplolepis spinosa, described by Bagatto and Shorthouse (1991). alters the nutritional status of attacked stems. Values for Ca, Mg, and Mn in the leaves Diplolepis spinosa is a small, rufous-or- and stems of the control samples (Table 1) ange cynipid wasp, the larvae of which in- are considerably lower than those reported duce a large, spiny, multi-chambered stem in the literature (Carlson and Bergman, 1966; gall (Fig. 1). Like all gall insects, D. spinosa Diaz et al., 1985; Johansson, 1978); how- receives food and shelter from the host plant ever, the study plants appeared healthy and as the larvae stimulate and then control wound were producing numerous blossoms. Values or defensive reactions of attacked organs. The for the remaining nutrients in the nongalled Fig. 1. Mature gall on stem of Rosa rugosa. Note gall undergoes three phases of development stems fall within the range considered ade- stunted leaves associated with the gall. referred to as initiation, growth, and matu- ration (Rohfritsch and Shorthouse, 1982). In most cases, the gall prevents further stem growth; however, stunted, chlorotic leaves are commonly associated with the gall op- posite the point of attachment (Fig. 1). On four occasions we found flower buds asso- ciated with maturing galls, and in all cases the buds were stunted and failed to open. In Aug. 1989 we collected four normal unbranched sucker shoot stems of Rosa ru- gosa and four sucker shoot stems with typ-

Received for publication 28 Dec. 1990. This re- search was supported by grants from the Natural Sciences and Engineering Research Council of Canada, the Laurentian Univ. Research Fund, and the American Rose Society. The cost of publish- ing this paper was defrayed in part by the payment of page charges. Under postal regulations, this pa- per therefore must be hereby marked advertise- yReported tissue concentrations in nutritionally sufficient rose plants given in parentheses (Carlson and ment solely to indicate this fact. Bergman, 1966; Diaz et al., 1985; Johansson, 1978). x lTo whom reprint requests should be addressed. Means shown ± SE.

HORTSCIENCE, VOL. 26(10), OCTOBER 1991 1283 physiological sinks and support the conten- healthy plants (Cockfield, 1988), the possi- correlation with flower yield. Proc. Amer. Soc. tion that galls, and the insects within, drain bility exists that galled roses will be more Hort. Sci. 88:671-677. nutrients from their host plants (Abrahamson susceptible to attack by other phytophagous Cockfield, S.D. 1988. Relative availability of ni- and Weis, 1987; Andersen and Mizell, 1987). pests. trogen in host plants of invertebrate herbivores: Furthermore, the high concentrations of sev- Three possible nutritional and physiological eral nutrients in the galls and adjoining stem Literature Cited definitions. Oecologia 77:91-94. tissues suggest that D. spinosa mobilizes nu- Diaz, A., M. Fernandez-Falcon, V. Garcia, and Abrahamson, W.G. and A.E. Weis. 1987. The L Sosa. 1985. Study of nutrient distribution in trients from other parts of the host plant. nutritional ecology of gall-makers, p. the rose bush for foliar diagnostic purposes. However, for the rose grower, the results 235-238. In: F. Slansky and J.G. Rodriguez Agrochimica 29(5-6):428-434. provide evidence that cynipid galls stress their (eds.). The nutritional ecology of insects, mites hosts and should be controlled or removed. and spiders. Wiley, New York. Johansson, J. 1978. Effects of nutrient levels on This precaution could be of particular im- Andersen, P.C. and R.F. Mizell, III. 1987. Phys- growth, flowering and leaf nutrient content of portance when roses are growing in nutrient- iological effects of galls induced by Phylloxera greenhouse roses. Acta Agr. Scandinavia. deficient soils. notabilis (Homoptera: Phylloxeridae) on pecan 28:363-386. Future studies of this gall system and oth- foliage. Environ. Entomol. 16:264-268. Rohfritsch, O. and J.D. Shorthouse. 1982. ers should include an examination of growth Bagatto, G. and J.D. Shorthouse. 1991. Accu- galls. p. 131-152. In: G. Kahl and J.S. Schell rates of galled and nongalled stems, along mulation of copper and nickel in plant tissues (eds.). Molecular biology of plant tumours. Ac- and galls of Hemadas nubilipennis (Hymenop- ademic, New York. with a study of how galls alter the architec- tera: ) on lowbush blueberry near ture of their host plants. The results pre- Shorthouse, J.D. 1988. Occurrence of two gall an ore smelter at Sudbury, Ontario, Canada. wasps of the Diplolepis (Hymenoptera: sented here indicate that galled plants are Can. J. Bot. (In press.) Cynipidae) on the domestic shrub rose, Rosa stressed, and since stressed plants are com- Carlson, W.H. and E.L. Bergman. 1966. Tissue rugosa Thunb. (Rosaceae). Can. Entomol. monly more vulnerable to insect attack than analysis of greenhouse roses (Rosa hybrida) and 120:727-737.

1284 HORTSCIENCE, VOL. 26(10), OCTOBER 1991