Search

Sorghum [Sorghum bicolor (L.) Moench ‘Early Hegari’] seedlings were grown in nutrient solutions with various combinations of chlorides, nitrates, sulfates, and phosphates (dihydrogen) of ammonium, potassium, calcium, and magnesium to determine the effects of the mineral elements on hydrocyanic acid potential (HCN-p) in leaves. Growth was maximum or near maximum for the shoots of sorghum seedlings at 1 to 30 meq salt in solution and for the roots at 0.3 to 3 meq. The duration of exposure of seedlings to the treatment solution required for significant changes in HCN-p was at least 2 days. The effects of various concentrations of individual salts on HCN-p were: KCl, a slight decrease; K2SO4, NH4Cl, NH4NO3, (NH4)2SO4, NH4H2PO4, and KH2PO4)2, an increase; and KNO3, CaCl2, Ca(NO3)2, CaSO4, MgCl2, Mg(NO3)2, and MgSO4, no significant effect. When potassium and ammonium salts of a common anion were added in 3 × 3 factorial experiments, increasing levels of ammonium salts at the 1X potassium level generally increased HCN-p, but varied effects of higher levels of potassium salts were observed. In factorial experiments, different levels of pairs of potassium salts of the various anion had no significant effect on HCN-p. In experiments with pairs of ammonium salts, increases in salt concentration generally increased HCN-p. It was concluded that large changes in salt concentration affected alterations in HCN-p were associated with changes in ammonium salt levels

The ch4 mutant of sweetclover (Melilotus alba) has previously been demonstrated to be partially deficient in chlorophyll and to have a higher ratio of chlorophyll a to b than normal plants. We were able to substantiate these findings when plants were grown at 23 degrees C and lower (permissive temperatures). However, when grown at 26 degrees C (nonpermissive temperature) the plants produced small yellow leaves which exhibited one-twentieth the chlorophyll content of normal plants. Affected leaves did not increase their chlorophyll content when plants were incubated at permissive temperatures, but leaves which developed at the lower temperature contained increased amounts of chlorophyll. Similarly, only new leaves, not previously grown leaves, exhibited the yellow phenotype when the mutant plant was shifted from the permissive temperature to the nonpermissive temperature. Ribulose 1,5-bisphosphate carboxylase activity was decreased by half, relative to normal plants, in the mutant plants grown at the nonpermissive temperature, indicating that general protein synthesis was not greatly impaired and that the effect of the mutation was perhaps specific for chlorophyll content. HPLC analysis indicated that carotenoid content was not diminished to the same extent as chlorophyll and we have determined that the thylakoid protein kinase is not altered, as is the case for other chlorophyll b-deficient mutants. Experiments suggest that changes in photoperiod may be able to modulate the effect of temperature.

In a study of seedlings of 72 entries representing 39 species, 14 genera, and 2 tribes of grasses, only entries of the genus Sorghastrum were found to contain the cyanogenic glucoside, dhurrin [(S)-p-hydroxymandelonitrile β-D-glucopyranoside], Dhurrin was found in seedlings of each of the 10 entries of indiangrass [Sorghastrum nutans (L.) Nash] and the three entries of S. pellitum. Seedlings of five indiangrass cultivars were assayed for dhurrin concentration. Levels expressed as HCN potential (HCN-p) ranged from about 900 ppm for ‘Holt’ to about 1,200 ppm for ‘Llano’ and ‘Oto’.Further studies with Oto seedlings indicated that first leaves were higher and more uniform in HCN-p than were other portions of the shoot, but all shoot portions contained dhurrin. Seedling roots also contained dhurrin, but the HCN-p of roots was appreciably lower than that of shoots. Spectral evidence indicated that dhurrin was present in extracts of leaves of tillers harvested from field-grown plants. Dhurrin was not detected in all such tillers, however. Indiangrass tiller extracts contained more interfering substances than seedling extracts, in agreement with previous observations on Sorghum seedlings

Secondary constrictions were determined in one chromosome pair in the complements of M. intesta, M. macrocarpa, M. italica (subgenus Micromeli/otus), and M. alba and M. officinalis (subgenus Meli/otus). The karyotypes of M. intesta and M. macrocarpa were found to be similar. Chromosomes of M. italica were larger than the chromosomes of the other four Melilotus species. The chromosome size in M. italica suggests the presence of large chromosomes in an ancestral or pro-Melilotus prototype. The chromosomes with satellites of M. intesta and M. macrocarpa appear to differ in morphology from the satellite chromosomes of M. alba and M. officinalis by a paracentric inversion. The morphology of the satellite chromosomes in M. italica is thought to rep resent a more primitive type than in the other Meli/otus species studied .

When dhurrin [p-hydroxy-(S)-mandelonitrile-tJ-D-glucoside], the cyanogenic glucoside of sorghum [Sorghum bicolor (L.) Moench], is hydrolyzed by autoclaving, p-hydroxybenzaldehyde (P-HB) is released. The spectrophotometric determination of pHB concentration in autoclaved sorghum leaf extracts provides a measure of the hydrocyanic acid potential (HCN-p) of leaf tissue. Extracts of field-grown sorghum leaves contained substances that interfered with this procedure, but ether extraction effectively separated p-HB from these interfering materials. We observed that when flag leaf tissue from field-grown sorghum was dried at 75°C and then autoclaved, HCN-p values were about three times as high as those based on tissue that was autoclaved without drying. Investigations of this apparent enhancement supported the conclusion that when fresh field-grown sorghum leaf tissue was autoclaved, dhurrin was extensively altered or lost, but neither p-HB nor HCN was produced. Drying the tissue at 75°C prior to autoclaving effectively reduced this loss. Inclusion of tissue drying and ether extraction steps in the spectrophotometric assay made this procedure, which was designed for use with sorghum seedlings, satisfactory for use with field-grown sorghum leaves. Additional index words: Cyanogenesis, Dhurrin, p-Hydroxybenzaldehyde, Prussic acid, Sorghum bicolor (L.) Moench, Spectrophotometric assay. AN earlier paper from this laboratory described a procedure for the spectrophotometric assay of the hydrocyanic acid potential (HCN-p) of young chamber-grown sorghum [Sorghum bicolor (L.) Moench] seedlings (2). The procedure was based on determination of p-hydroxybenzaldehyde (P-HB) released when dhurrin [p-hydroxy-(S)-mandelonitrileI3-D-glucoside], the cyanogenic compound of sorghum, was autoclaved in water. The absorbance spectrum of p-HB in alkaline solution has a pronounced peak at 330 nm. Autoclaved extracts of leaves from young seedlings displayed absorbance spectra very similar to that of pure p-HB (2); for such leaves A3 30 values of extracts diluted in base, provided a reliable measure of HCN-p (2,6). This seedling assay procedure was used successfully in a program of divergent selection for HCN-p in sudangrass [So sudanense (Piper) Stapf] (3). Absorbance spectra of extracts from young fieldgrown sorghum tillers differed appreciably from that of pure p-HB (6). Spectra of some tiller extracts lacked a 330-nm peak; those of other extracts had peaks at 330 nm, but their shape suggested extensive non-pHB absorbance at 330 nm. Reliability of the spectrophotometric procedure for use with tillers was improved by fractionating the autoclaved tiller extracts with ether. The p-HB in these extracts was soluble in the ether phase whereas most of the interfering materials remained in the aqueous phase (5). We have observed that when extracts obtained by autoclaving fresh field-grown sorghum flag leaves were diluted in base and scanned, the resulting spectra lacked well defined peaks at 330 nm. However, spectra obtained from oven-dried samples had well 1158 defined 330-nm peaks. The objective of this study was to investigate the effects of leaf drying and other treatments on the spectra of leaf extracts and on the HCN-p values based on these spectra. In the course of these experiments, the spectrophotometric procedure for seedlings (2) was modified, making it satisfactory for use with field-grown sorghum leaves. MATERIALS AND METHODS

The KSS and N32 sorghum [Sorghum bieolor (L.) Moench) lines are low and high, respectively, in the hydrocyanic acid potential (HCN-p) of mature leaves. This difference is conditioned primarily by a single pair of alleles. The main objective of this study was to determine, at various stages of plant growth and various times during the growing season, the HCN-p of upper leaves and tillers of fieldgrown plants of these two parental lines and of two low-HCN-p F3 lines derived from crosses between KS8 and N32. The four entries were grown in a randomized complete block design with three replications in 1985. Samples of leaf tissue were dried, ground, and extracted, and cyanide in the extracts was assayed colorimetrically. Using a mean HCN-p level of SOO mg kgI dry wt to separate safe from unsafe sorghum forage, all samples of KS8 mature leaves and tillers would be considered safe, and all N32 samples would be considered potentially dangerous. Values for most of the samples of the FJ lines fell within the safe range, but some samples of young regrowth exceeded the 500 mg kg-I limit. Regressions of HCN-p on height for upper leaves of main stems and of tillers indicated a significant negative relationship for all entries except for leaves from the main stems of KS8. However, the relationship was not close enough to support the use of plant or tiller height as a reliable indicator of HCN-p. Levels of HCN-p also were determined for mature leaves and young regrowth of hybrids involving KS8, N32, and 'Redlan' sorghums as seed parents and NP25, 'Piper,' and 'Greenlear sudangrasses [So sudanense (Piper) Stapf) as pollinators. Results indicated that for minimizing the risk of cyanide poisoning, KS8 would be the seed parent of choice, and NP2S and Piper would be the preferred pollinators. Additional index words: Cyanogenesis, Prussic acid, Sorghum bicolor (L.) Moench, Sorghum sudanense (Piper) Stapf, Sorghum X sudangrass hybrids. A RECENT COMPARISON of the hydrocyanic acid potential (HCN-p) of the grain sorghum [Sorghum bicolor (L.) Moench] lines, KS8 and N32, indicated that seedling leaves of both lines were high in HCNp, but mature leaves from field-grown plants differed greatly, flag leaves of N32 being at least 10 times as high in HCN-p as KS8 flag leaves (3). The large difference in HCN-p between KS8 and N32 was detected in field-grown plants within about 5 weeks after planting (5), and was found to be conditioned primarily by a single pair of alleles (3). The objective of the present study was to determine, at various times during the growing season and at various stages of plant development, the HCN-p of upper leaves and tillers of KS8, N32, and two low-HCN-p F3lines derived from crosses 'between KS8 and N32. Determinations of this type are needed for appropriate management decisions concerning the safety of sorghum forage for grazing liveI Contribution from the USDA-ARS and the Nebraska Agric. Res. :Div., Lincoln, NE 68583. Published as Paper no. 8101, Journal Series, Nebraska A$ric. Res. Div. Research was conducted under Project 12-114. Received 31 July 1986. 2 George Holmes professor of agronomy; supervisory research ge~eticist, USDA-ARS; and research geneticist, USDA-ARS. respectively. Published in Crop Sci. 27:903-906 (1987). 903 stock. It would be unusual for farmers to use grain sorghums such as KS8 and N32 as pasture except during the period after grain harvest. However, sudangrass [So sudanense (Piper) Stapf] and sorghum X sudangrass hybrids are often used for pasture and greenchop, and the HCN-p of these forages at various growth stages is important to the livestock producer. Therefore, several other sorghum and sudangrass lines and their FI hybrids also were included in this study. MATERIALS AND METHODS

Fresh leaves of deer’s tongue contain large quantities (more than 10% of the dry weight, in some cases) of o-hydroxycinnamic acid (o-HCA). Bothcis- andtrans-o- HCA are present, and both isomers exist in the fresh tissue predominantly as glucosides. Cured deer’s tongue leaves contain relatively high levels of coumarin and lower amounts of o-HCA glucosides. It is probable that during the curing processcis-o-HCA glucoside is hydrolyzed by an endogenous s-glucosidase, and that the liberated cis-o-HCA lactonizes spontaneously to form coumarin.

Seeds of barley, maize, mustard, and safflower were treated with various doses of x rays and thermal neutrons, and observations of height, survival, and pollen abnormalities were made on the plants resulting from these seeds. In most cases a close relationship was found to exist between the three criteria of irradiation effect and dose of x rays or thermal neutrons. However, it is necessary that care be used in selecting the age of plant at which height and survival data are taken. The results suggest that although the chromosomes of the four species may differ widely in sensitivity to radiation damage, within each type of radiation the four species appear very similar with respect to influence on plant height per chromosomal abnormality. (auth)

Information obtained from the F" F., and Fa generations of crosses between the normal annual M. alba progenitor line and the four mutants, short-internode dwarf, curled leaf, multifoliolate leaf, and cotyledonary branching, indicates that each character is conditioned by a single pair of alleles. The multifoliolate leaf character is dominant over the normal phenotype; the other three mutant characters are recessive. The symbols dw, cl, Mj, and cb are proposed as designations for the respective mutant genes. ----------------Additional index word: Sweetclover. INHERITANCE studies in white-flowered sweetclover, Melilotus alba Desr., have dealt with a relatively small number of mutations, most of which occurred spontaneously. The 1965 review of Smith and Gorz (10) lists only 19 mutant traits for which the mode of inheritance has been elucidated. Since 1965 an additional mutant gene, responsible for the unifoliolate leaf-cauliflower inflorescence phenotype, has been reported by Goplen (3). The need for added genetic markers in this species is obvious. Within the past 15 years several workers have succeeded in isolating new mutant strains of M. alba following the use of ionizing radiation or chemical mutagens. Mutagenic work on this species was started in the early 1950's when Micke (7) began a series of experiments in which dry seeds were irradiated with X rays or thermal neutrons. From this and other studies of Micke and of Scheibe and Hiilsmann (8), who utilized chemical mutagens, some 690 independently occurring mutations had been identified by 1966 (9). Very few of these mutants have been subjected to gene

The ashgray blister beetle, Epicauta fabricii (LeConte); the striped blister beetle, E. vittata (F.) ; and the margined blister beetle, E. pestifera Werner, collected in the field from plants of sweetclover low in coumarin content, fed preferentially in the laboratory on low-coumarin leaves when offered a choice between leaves from high-and low-coumarin plants. The black blister beetle, E. pennsylvanica (De Geer), not known to feed on sweet-clover foliage, also fed preferentially in the laboratory on leaves from low-coumarin plants, although the rate of feeding was much slower than for the other species. By using appropriate genotypes of sweetclover and controlling the type of light to which the plants were exposed, it was possible to offer the beetles choices among leaves differing greatly in their content of specific coumarin-related compounds. Both cis-o -hydroxycinnamic acid ( cis-o -HCA) glucoside and coumarin (the lactone form of cis-oo -HCA) were strong feeding deterrents; trans-o -HCA glucoside had no significant deterring effect.

Seedlings of both KS8 and N32 sorghum [Sorghum bicolor (L.) Moench] were high in dhurrin [p-hydroxy-(S)-mandelonitrile-pD'glucoside] and thus in hydrocyanic acid potential (HCN-p), but the HCN-p of mature leaves from field-grown plants of KS8 was only about one-tenth as high as that of N32. A study of the inheritance of this large difference between KS8 and N32 revealed that a single major gene pair was responsible. There were no obvious maternal effects and F,'s were generally intermediate in HCN-p level between the two parents, indicating that neither high nor low HCN-p was completely dominant. HCN-p level was influenced by genetic background since mean values of the low and intermediate HCN-p classes increased slightly as the proportion of genetic background ascribable to N32 was increased. A survey of 102 additional sorghums revealed that the gene for low HCN-p carried by KS8 does not appear to occur widely because KS8 had the lowest HCN-p of all entries assayed. Additional index words: Cyanogenesis, Genetics, Hydrocyanic acid,p-Hydroxybenzaldehyde, Prussic acid, Sorghum bicolor (L.) Moench, Spectrophotometric assay. THE biosynthesis of the cyanogenic glucoside, dhurrin [p-hydroxy-(S)-mandelonitrile-l3-D-glucoside], in sorghum [Sorghum bicolor (L.) Moench] has been studied extensively, and much has been learned about the process (I). However, studies of the inheritance of dhurrin content in this species have been much less conclusive. Nass (10) reviewed published reports on the inheritance of cyanogenesis in sorghum, various Lotus species, and Trifolium repens 1., and concluded that the situation in sorghum was more complex than that in other species. There was general agreement that the hydrocyanic acid potential (HCN-p) of sorghum leaves was a heritable trait, but the reports failed to agree on such matters as dominance of low or high HCN-p and the number of genes involved. Lack of agreement may have been due in part to differences among the various studies with respect to lines and cultivars used, conditions of growth and sampling, and analytical procedures. Krauss (7) concluded that HCN-p in sorghum was governed by four gene pairs with additive effects and without dominance. His work was based on crosses among four sorghum cultivars differing in HCN-p. More recent reports have included information about the HCN-p of hybrids and their parents in forage sorghum (13, 15) and also in sudangrass and sudangrass-sorghum combinations (14), but these studies were not designed specifically to investigate the inheritance of HCN-p. Recently Haskins et al. (4) reported that the two sorghum lines, KS8 and N32, both had high HCNP levels as seedlings, but when upper leaves from , Contribution from the USDA-ARS and the Nebraska Agric. Res. Div., Lincoln, NE 68583. Published as Paper no. 7724, JournalSeries, Nebraska Agric. Res. Div. The work reported wasconducted under Nebraska Agric. Res. Div. Project 12-114. Received 27 Mar. 1985. 2 Supervisory research geneticist, USDA-ARS; George Holmes professor of agronomy; ana supervisoryresearch geneticist, USDAARS, respectively. 65 field-grown plants were compared in mid-August, the HCN-p of KS8 was only about one-tenth as high as that of N32. The existence of this large difference suggested that these lines might be useful in a study of the inheritance of HCN-p in mature sorghum leaves. MATERIALS AND METHODS

The tests utilize small samples of leaf tissue crushed on sheets of filter paper. For detection of o-hydroxycinnamic acid (o-HCA), a crude preparation containing sweetdover ,B-glucosidase is added to a spot of crushed tissue to insure hydrolysis of glucosidically bound o-HCA. An extract containing cis-o-HCA glucoside is added to another tissue spot to serve as the substrate for the detection of ,B-glucosidase activity. Finally, all spots are moistened with NaOH and are scored foJ:' fluorescence under ultraviolet light. Results of qualitative and quantitative tests on plants representing various genotypes are compared_ The importance of exposing plants to sunlight before testing for o-HCA is discussed. The tests are particularly useful in genetic studies, in the development and maintenance of breeding lines, and in testing for contamination in low-o-HCA sweetclover varieties. Additional index words: Sweetc1over, Coumarin. S qualitative and quantitative fluorometric procedures for the assay of coumarin and related compounds in sweetclover (Melilotus alba Desr.) are discussed in the review of Smith and Gorz (14). Of the qualitative procedures cited in that review, the most recent were published in 1958 (2, 12). Biosynthe tic studies published since 1958 have contributed greatly to an improved understanding of such details 1 Contribution from the Crops Research Division, Agricultural Research Service, U. S. Department of Agriculture, and the Nebraska Agricultural Experiment Station, Lincoln, Neb. Supported in part by the National Science Foundation (Grant Nos. GB-1l48 and GB-8280). Published with the approval of the Director as Paper No. 2720, Journal Series, Nebraska Agr. Exp. Sta. Received Dec. 12, 1969. 2 Bert Rodgers Professor of Agronomy, University of Nebraska, and Research Geneticist, Crops Research Division, ARS, USDA, Lincoln, Neb. The technical assistance of Mrs. Linda Helton, David Gilchrist, and Ralph Kulm is gratefully acknowledged. of the coumarin pathway as the relationship of the glucosides of cisand trans-o-hydroxycinnamic acid (o-HCA) to free and bound coumarin, the role of the enzyme, fJ-glucosidase, in the pathway, and the control of o-HCA level and fJ-glucosidase activity, respectively, by the Cujcu and Bjb allelic pairs (5). This understanding, in turn, has permitted the development of improved qualitative tests for both o-HCA and fJglucosidase. In this report, the modified tests currently in use in this laboratory are described, and results of their application are presented and discussed. Sweetclover plants of the CuCu genotype are relatively high in content of the glucosides of cisand trans-o-HCA; Cucu plants are intermediate; and cucu plants are low in content (3, II). Similarly, preparations of BB plants are high in fJ-glucosidase activity, those of Bb plants are intermediate, and bb preparations appear to be inactive (6). Experience has shown that the qualitative tests for o-HCA content and fJglucosidase activity are highly reliable in cases involving plants homozygous for the Cujcu and Bjb genes, but that some difficulty in classification may be expected if the plant population includes individuals heterozygous for either or both of the two genes. For this reason, plants of all possible genotypes with respect to these two allelic pairs were included

Six ethyl methanesulfonate-induced mutants of Melilotus alba Desr. were studied. Five of the mutants behaved as monogenic recessives. In the sixth mutant, two independent recessive alleles were responsible for the observed phenotype. Of these two genes, one had a phenotypic effect similar to the five single-gene mutants. The other (the veined gene) caused a chlorophyll deficiency in which the leaf veins were darker in color than the tissue between the veins. Complementation analysis revealed that five of the seven genes detected iu the mutants were nonallelic. Suggested designations for the five genes are ch" ch6, ch., ch7, and chv. Additional index words: Sweetc1over, Ethyl methanesulfonate. G ENE mutations influencing the green coloration of photosynthetically active parts are among the most common spontaneous or induced alterations arising in higher plants. These are usually referred to collectively as chlorophyll mutants. They range from lethal to semilethal or completely viable types having white, yellow, or pale green leaves and stems. Many of the lethal mutants would be of great value in biosynthetic studies, but only rarely is a lethal mutant found whose defect can be corrected by altering the cultural conditions. Consequently, most investigations of chlorophyll mutants in higher plants have utilized nonlethal types capable of growth and reproduction when planted in soil in a growth chamber, greenhouse, or in the field. Knowledge of the 1 Contribution from the Crops Research Division, Agricultural Research Service, U. S. Department of Agriculture, and the Ne· braska Agricultural Experiment Station, Lincoln, Nebraska. Sup· ported in part by the National Science Foundation (Grants Nos. GB·1l48 and GB-8280). Published with the approval of the Di· rector as Paper No. 2639, Journal Series, Nebraska Agr. Exp. Sta. Most of the data were taken from a thesis submitted by the senior author to the University of Nebraska in partial fulfill· ment of the requirements for the M.S. degree. Received Sept. 11, 1969. 2 Formerly graduate assistant in Agronomy, University of Nebraska (now graduate student in the Agronomy Department, University of Illinois); Research Geneticist, Crops Research Di· vision, ARS, USDA; and Bert Rodgers Professor of Agronomy, University of Nebraska, Lincoln, Nebr., 68503, respectively. GEGENBACH ET AL.: INDlJCED MUTANTS IN MEULOTUS ALBA 155 inheritance of such mutants enhances their usefulness in biochemical studies. Interspecific hyhridization within the genus Melilotus results in many chlorophyll-deficient hybrids (6), but most of these have not been studied genetically. The only hybrid subjected to detailed genetic analysis arose from the cross of .lvI. alba Desr. and 1.'11. dentala (Waldst. & Kit.) Pel's. (7). Chlorophyll-deficient Fl hy~r~ds were preserved by grafting to normal M. ottzcznalis (L.) Lam. plants, and were used as the female parents in back crosses with }vL alba. Effects of several nonallelic genes controlling chlorophyll deficiency were noted in progenies of the resulting backcross plants. Three of these genes, identified as chI> Ch2' and Ch3, produced chlorophyll-deficient plants when homozygous, as well as in certain heterozygous combinations. The triply heterozygous condition was lethal in the seedling stage. Subsequent work suggested that chlorophyll deficiency in the Fl hybrid of the M. alba X !vI. dentala cross resulted from the interaction hetween nonallelic genes of the two species (7). Clarke (2) described two spontaneous chlorophylldeficient mutants of 1'11. alba that were semilethal in the greenhouse, but almost 100% of the mutant plants died under field conditions. Each mutant was inherited as an independent monogenic recessive resulting in the production of pale green seedlings. The two genes were designated Pgl and Pg2. A.rece~t report from this laboratory (4) deals with the lsolatIOn of several types of mutants in an annual strain of M. alba, following seed treatments with ethyl methanesulfonate. Numerous chlorophyll-deficient mutants were detected. Many of the deficiencies were lethal, but some mutant lines could be successfully propagated in growth chambers or the greenhouse. Six of these viable, chlorophyll-deficient mutants were available for study when the present work was initiated. The studies reported were conducted to determine the mode of inheritance of these six mutants and to inv~stigate the possihility of allelism among the genes lIlvolved. MATERIALS AND METHODS The chlorophyll-deficient lines were derived from the same source and grown under essentially the same conditions as the morphological mutants previously described (3). All parental, Fj , and F" plants were grown in growth chambers while the F2 progenies were grown in the greenhouse. Characteristics of the parental lines are given in Table 1. The system used for color classification was as follows: #1 albino, lethal; #2 yellow, may be lethal; #3 yellow-green; #4 light green; #5 dark (normal) green. The letter "v" associa ted with a color class denotes the presence of a prominent leaf venation which was considerably darker in color, particularly in the early stages of growth, than the leaves of the #3 or #4 mntants in which it appeared. The veined mutant was readily distinguished from all other mutants at all stages of growth in the growth chambers as well as the greenhouse. The five non veined mutants could Table 1. Characteristics of six chlorophyll·deficient mutant lines and the normal green control. Color Height at Seed Line class Color maturity. em pr~ Q839 ~ Yellow-green ;:]0-.18 Good Q84,1 4 Light green 1.'5-25 l)oor QS44 4 Light green 20-30 eoad Q851 4 Light green :W-::lS Good Q856 4 Ught greer: :~0-:~8 Fair Q858 :1V Yellow-green and vC'inC'c1 80-88 Good Q525 ., ~onn!ll green ~l5-50 Fx('t'llt'nt --------~---.-------."-----------"-be distinguished from each other only during the first 2 to 3 weeks of growth in the growth chamber. All mutants were easily distinguished from the normal control at all stages of growth. Th.e crossing procedure was identical to that previously described (3) except that an attempt was made to obtain seed from all normal X mutant crosses and their reciprocals. For complementation studies, crosses among the six mutants were made in all possible comhinations. but no attempt was made to obtain all possible reciprocal crosses. The 1', plants obtained from each cross were compared to plants of the parental lines with respect to leaf color and venation, height, general vigor, and seed set. For each mutant, 100 to 200 1'2 seeds, obtained by self-pollination of one to three F, plants, were planted to obtain F" segregation ratios. Data from F" progenies segregating for the same phenotype were pooled. The progenies of several F2 plants classified as recessive were planted to determine whether the recessive condition bred true, and the progenies of at least 19 1'. plants of the dominant phenotype also were checked for segregation. In this progeny testing, either 18 or 27 seeds of each dominant F. plant were used, depending on whether the F3 families would be expected to scgrf~gate into two or four classes, respectively. RESULTS AND DISCUSSION Crosses Between Normal and Mutant Lines F t plants were obtained from 10 of the 12 possible reciprocal combinations of the six mutant lines crossed with normal plants. All Fl plants were phenotypically similar to the normal parent in leaf coloration, plant height, and vigor indicating that all six mutants were conditioned by recessive genes. Ratios of normal to chlorophyll-deficient plants in the F 2 and segregating F3 progenies from five of the six mutant lines were approximately 3: I (Table 2), suggesting a monogenic inheritance with normal leaf color completely dominant over each chlorophyll-deficient mutant. Phenotypically mutant F2 plants from each of these five lines bred true in the F3 generation. F2 and some Fa progenies from the sixth mutant line, Q8.58, segregated into four distinct classes as follows: dark green with normal veins (#5), light green with normal veins (#4), light green with dark green veins (#4v), and yellowgreen with light green veins (#3v). The F2 ratio observed gave a satisfactory fit to a 9:3:3:1 ratio, indicating that two recessive genes were apparently responsible for the chlorophyll deficiency and venation observed in line Q858. The conclusion is reasonable that plants of color class #4 resulted when one of the gene pairs was in the homozygous recessive condition; #4v plants were produced when the other gene pair Table 2_ Chi-square analyses of F2 and F3 segregating populations from crosses of six chlorophyll-deficient mutants to normal plants. :\tutant Gener_ ~o. of Distribution among color classes line ation* families #,5 #4v #4 #3v #3 Q839 F~ (5) 3 1,53 46 F, (,5) 14 149 41 Q843 F, (,5) 132 44 r30) 1,1 118 40 QR4·1 Fz (.') 141 51 F~ (.'») 1.5 149 :il Qa,51 Fz (5) 81 20 }~ (.'5) 1.1 167 D1 QSS6 r~ (5) 76 22 F~ (5) 11 10:3 2G Q8.'58 F, (5) 57 20 22 Fj (.5) S.'5 2.s F3 (.'5) 2 2·j 8 F',,(.,) ]2 12:1 40 29 10 1-'" (4v) 1.5 1:l8 49 F; (4) .')1 1,8 of-flt to :l 9

A bioassay employing sweetclover root disks impregnated with various plant extracts has been developed. The bioassay has been used to demonstrate the distribution of substances influencing feeding in fractionated water-methanol-chloroform extracts of Melilotus infesta Guss. and M. officinalis L. Lam. leaves. Indications are that substances responsible for the resistance of M. officinalis a or the susceptibility of M. officinalis to feeding by adult sweetclover weevils, Sitona cylindricollis , Fahraeus, reside in the water-methanol fraction.

Flag leaves of 'Colman' forage sorghum (Sorghum bicolor) contain at least 25 times as much leucoanthocyanidin (LAC) and approximately half as much of the cyanogenic glucoside, dhurrin, as do flag leaves of 'White Collier' forage sorghum. Assays of flag leaves from 119 F2 plants and 11 F5 lines from crosses between these two cultivars revealed a statistically significant negative association between levels of LAC and dhurrin. Both LAC and dhurrin are aromatic compounds, and the negative association between the two may be the result of competition for intermediates or products of the aromatic biosynthetic pathway. This rationale appears to be quite different from that for the negative association reported for levels of tannin and cyanide in Lotus corniculatus. Although the negative relationship between LAC and dhurrin in sorghum was statistically significant, the association was not consistent enough to suggest that either trait could be used reliably in selecting or breeding to modify the other trait.

Phenylalanine ammonia-lyase activity was investigated in preparations representing various parts of sweetclover (Melilotus alba Desr.) plants of CuCu and cucu genotypes. In contrast to other plant parts, very young leaves and stems of CuCu plants displayed high phenylalanine ammonia-lyase activity. Initial leaf samples from CuCu plants were approximately 3 times as high in enzyme activity as leaves from cucu plants, but stems were only slightly higher in activity. Defoliation of the plants resulted in decreased enzyme activity, increased o-hydroxycinnamic acid content, and essentially no difference in enzyme activity between the genotypes. It appears that phenylalanine ammonia-lyase activity in leaves is not primarily controlled by the Cu/cu alleles and that the reaction catalyzed by this enzyme is not the limiting step in o-hydroxycinnamic acid synthesis.

The beta-glucosides of coumarinic and o-coumaric acids were detected in extracts of cotyledons, exocarps, and seedling leaves of the tonka bean. The existence of these compounds and the presence of a beta-glucosidase having specificity for coumarinyl glucoside suggest that the tonka bean synthesizes coumarin by a pathway similar to the one found in sweetclover.

Deterrent B, a compound apparently involved in the resistance of Melilotus infesta to the sweetclover weevil Sitona cylindricollis, has been isolated from leaves by a combination of preparative paper chromatography, sublimation, and crystallization. The compound has been identified as ammonium nitrate. Isolated deterrent B and ammonium nitrate have identical feeding deterrent activities. Although the deterrent principle was isolated as the ammonium salt, nitrate ion is probably responsible for the deterrent activity in vivo.

Week-old shoots of 50 Sorghum entries representing 22 species, plus four Sorghum entries of undesignated species, were dried at 75° and the dried tissue extracted with water at room temperature. The resulting extracts were diluted in 0.1 M sodium hydroxide and spectra were scanned immediately to provide a measure of free p -hydroxybenzaldehyde. Scans were repeated after the basic solutions had stood for 3 hr at room temperature to permit hydrolysis of dhurrin ( S - p -hydroxymandelonitrile β- D -glucopyranoside). Without exception, the quantity of free p -hydroxybenzaldehyde was very small in relation to the quantity released by dhurrin hydrolysis.

Flag leaves of KS8 sorghum (Sorghum bicolor (L.) Moench] are low in dhurrin (p-hydroxy-(S)-mandelonitrile-p-D-glucosidel content and thus in hydrocyanic acid potential (HCN-p), and they contain little (if any) leucoanthocyanidin (LAC). Comparable leaves of 'Colman' sorghum are intermediate in HCN-p and high in LAC. This study was conducted to investigate the inheritance of HCN-p and LAC in crosses of KS8 X Colman. Flag leaves from field-grown plants of both parents; the Flo KS8 X Colman; the backcross, KS8 X (KS8 X Colman); and the F 2, (KS8 X Colman) selfed, were assayed for both HCN-p and LAC. Assays for HCN-p indicated that the backcross values provided a good fit to 1 intermediate:l low (Xl = 0.05, P = 0.82) and F2 results to 3 intermediate:l low (Xl = 0.96, P =0.33). For LAC, backcross results were 1 LAC +:1 LAC(Xl = 0.20, P = 0.65) and F2 values were 3 LAC+:1 LAC(exact fit). Classification for both traits yielded good fits to 1:1:1:1 for the backcross (Xl = 0.30, P = 0.96) and 9:3:3:1 for the F 2 (Xl = 3.24, P = 0.36). These results indicated that the difference in HCN-p between Colman and KS8 depended primarily on a single gene, the difference in LAC depended largely on a separate single gene, and the HCN-p gene and the LAC gene were not linked. A 1986 REPORT (1) indicated that a single gene difference was primarily responsible for the difference in dhurrin content (expressed as hydrocyanic acid potential, HCN-p) of mature leaves between KS8 (lowHCN-p) and N32 (high-HCN-p) lines of sorghum. Similarly, the large difference in flag-leaf leucoanthocyanidin (LAC) content between 'White Collier' (LAC-) and Colman (LAC+) sorghums depended largely on a single pair of genes (2). Both KS8 and N32 were LAC-, and the HCN-p values for flag leaves of both Colman and White Collier were intermediate between the values for KS8 and N32 (EA. Haskins and H.J. Gorz, 1988, unpublished observations). Therefore, the crosses used in the previous studies were considered inappropriate for investigating the possible linkage of the HCN-p and LAC traits. The objective of the present study was to investigate the segregation of these two traits in a cross of KS8 (low-HCN-p, LAC-) X Colman (intermediate-HCN-p, LAC+). Materials and Methods Colman (male-fertile), AKS8 (cytoplasmic male-sterile), and BKS8(male-fertile, sterility maintainer) sorghum plants weregrown at the Agronomy Farm, Lincoln, NE, during the summer of 1985. The status of Colman with respect to fertilityrestoration was unknown; therefore, to ensure that malefertile F, plants would be produced, BKS8 plants were hand emasculatedand pollinated with Colman pollen. In NovemFA Haskins, Dep, of Agronomy, and H.J. Gorz, USDA-ARS and Dep. ofAgronomy, Nebraska Agric. Res. Div.,Lincoln, NE 68583. Published as Paper no. 8484, Journal Series, Nebraska Agric. Res. Div. Research was conducted under Project no. 12-114. Received 27 Nov. 1987. *Corresponding author. Published in Crop Sci. 28:864-865 (1988) ber, 1985, the F, seed was sent to Puerto Rico (Tropical Agricultural Research Station, USDA-ARS, Mayaguez)where F, plants were self-pollinated to produce F2 seed, and were also used as pollen parents in backcrosses to AKS8. AKS8 X Colman crosses also were made during the summer of 1985 to obtain larger quantities ofseed from which F , plants could be grown for sampling. Plants of BKS8, Colman, the F, (AKS8 X Colman), the backcross [AKS8 X (BKS8 X Colman)], and the F2 [(BKS8 X Colman) selfed] were started in the greenhouse during the spring of 1986 and were transplanted to the Agronomy Farm on 29 May. Plants were placed 0.61 m apart in rows with a 0.76-m spacing. The experiment was planted in four replications, with each replication including one 10-plant row of each parent and the F" two rows of the backcross, and five rows of the F2• Thus, the experiment was designed to include 40 plants of each parent and the F I , 80 backcross plants, and 200 F2 plants. Entries were assigned at random to the 10 rows in each replication. Insufficient plants of both parents and the F, were available for transplanting; therefore, seeded rows of BKS8, Colman and the F, were used in two, three, and one of the replications, respectively. On 30 July, when panicles were emerging from most plants, the blade of the flagleaf was harvested from each plant. One KS8 and 21 F2 plants were not sufficiently advanced to allow positive identification of the flag leaf; for these plants, the blade ofthe youngest leaf with a visible collar was harvested. Midribs were removed from the leaf blades, and the remaining tissue was dried at 70 to 75°C overnight. The dry tissue was ground through a I-mm screen and stored in plastic vials at -18°C prior to extraction for assay. Extraction and assay for HCN-p were conducted as described previously (I) except that dhurrin was hydrolyzed enzymatically rather than with NaOH. The enzyme preparation was an extract made by soaking defatted almond meal (Sigma Chemical Co., St. Louis, MO)' in distilled water (8 mg mL -I) and filtering the suspension through Whatman no. 1 filter paper. To hydrolyze dhurrin, I mL of this filtrate was added to I mL of leaf extract, and the mixture was incubated in a parafilm-capped tube at room temperature for 1.25 h. Following this incubation, 8 mL of 0.1 MNaOH was added, and a l-mL portion of the resulting solution was assayed colorimetrically as described previously (I). The procedure described by Haskins and Gorz (2) was used for extraction and assay of LAC. Absorbance at 540 nm was used as a measure of LAC content. Results and Discussion Segregation for HCN-p Mean HCN-p values for KS8 and Colman were 29 and 135 mg kg-I, respectively, and standard errors were such that these means appeared to be well separated (Table 1). However, ranges in HCN-p for these two parents overlapped slightly, which caused some uncertainty in the classification of backcross and F2 plants as either low (L) or intermediate (I) in HCNp. As shown in the table, with 50 mg kg-I as the dividing line between the classes, only one of 37 KS8 plants was classified as I-HCN-p and only two of 40 Colman plants were classified as L-HCN-p. With this dividing line, 41 of the backcross plants were I-HCNI Names ofproducts are included for the benefit of the reader and do not imply endorsement or preferential treatment by the USDA or the Univ. of Nebraska.

Chemical analysis of seeds and seedlings of the CC and cc genotypes in Melilotus alba indicated that these alleles affect flavonoid biosynthesis. The CC seed coats contained orientin and iso -orientin, which were absent in the cc seed coats. The pigment responsible for the red pigmentation of young seedlings of CC genotypes was a cyanidin glycoside. The embryos of seeds of both the CC and cc genotypes contained a flavonoid tentatively identified as a 6,8-di- C -pentosylapigenin. The observation that 3′,4′-dihydroxyflavonoids were absent in the cc genotype and that 4′-hydroxyflavonoids were present in both genotypes indicated that the C/c alleles controlled the 3′-hydroxylation of flavonoids. The C/c alleles did not, however, control 3′-hydroxylation of cinnamic acids since caffeic acid was detected in both genotypes.

Free p -hydroxybenzaldehyde was not present in appreciable quantity on the surface or in the interior of week-old Sorghum bicolor shoots that had been heated to inactivate hydrolytic enzymes, nor was p -hydroxybenzaldehyde detected in epicuticular wax of greenhouse-grown sorghum ca 4.5 months old.