Excess Pollen Research Articles

Low seed viability of a rare aster

The reproductive ecology of rare plants is often unknown, yet the persistence of most plant populations depends on successful interactions with pollinators and favourable environmental conditions. Sexual reproduction maintains genetic diversity within plant populations using pollinators to transport pollen grains among plants of the same species and producing seeds. We investigated the reproduction of Yermo xanthocephalus Dorn (Desert yellowhead), a perennial member of Asteraceae endemic to central Wyoming, USA, which grows in only two locations within 8 km of each other. Our objectives were to assess the pollination and seed-set of Y. xanthocephalus. We conducted seed-set experiments in both populations to measure self-pollination and estimate if pollinating insects limited seed production. We used vane traps and pan traps to capture pollinators, and we examined pollen carried on bees. Yermo xanthocephalus can self-pollinate, but seed-set was extremely low. The number of viable seeds produced in flowerheads pollinated by insects did not differ from those produced by capitula hand-pollinated with excess pollen, indicating that pollinators did not limit seed production that year. Pollen from Y. xanthocephalus was carried by nine bee genera, suggesting that no one specific insect pollinates this plant. Only 12% of ovules produced viable seeds in the main population and 0% were viable in the other population, suggesting that something beyond pollinators limited seed-set. We recommend continued research to address what is limiting seed production to advance the knowledge and management of this declining plant species.

Pollen limitation and flower abortion in a wind-pollinated, masting tree.

Pollen limitation is a key assumption of theories that explain mast seeding, which is common among wind-pollinated and woody plants. In particular, the pollen coupling hypothesis and pollination Moran effect hypothesis assume pollen limitation as a factor that synchronizes seed crops across individuals. The existence of pollen limitation has not, however, been unambiguously demonstrated in wind-pollinated, masting trees. We conducted a two-year pollen supplementation experiment on a masting oak species, Quercus lobata. Supplemental pollen increased acorn set in one year but not in the other, supporting the importance of pollen coupling and pollination Moran effect models of mast seeding. We also tracked the fate of female flowers over five years and found that the vast majority of flowers were aborted for reasons unrelated to pollination, even in the presence of excess pollen. Pollen limitation can reduce annual seed set in a wind-pollinated tree, but factors other than pollen limitation cause the majority of flower abortion.

Selection for a floral trait is not mediated by pollen receipt even though seed set in the population is pollen‐limited

SummarySeed production in natural populations is often limited by quantity of pollen received. This pollen limitation has the potential to generate natural selection through female function favouring certain floral traits (hereafter ‘pollen‐mediated’ selection). Floral traits can, however, also be under selection via other mechanisms, and the influence of a particular trait on pollen limitation of individuals is rarely quantified.We took a trait‐based approach to pollen limitation by adding a quantitative trait to a previously published model. The modified model predicted impacts of both pollen supplementation and pollen reduction in seed production when the trait influences pollen receipt, and when the trait influences other model parameters. We then manipulated pollen availability in a population of the alpine New Zealand herbWahlenbergia albomarginatavarying in petal colour to test a long‐standing hypothesis that pale flower colour in that habitat is selected for via pollinators.Under pollen‐mediated selection, supplementing pollination should eliminate fitness dependence on a trait, while reducing pollination could exacerbate it. In contrast, if selection through female function on a floral trait is not solely pollen‐mediated, seed production should change with the trait value even when excess pollen is added. Furthermore, if the effect is on maximum seed production, the trait effect should increase with higher pollen availability as greater resources can be used for filling seeds from more fertilised ovules.InWahlenbergia albomarginata, average seed production was strongly pollen‐limited, with supplemental pollen increasing seeds by 43%. Selection through female function also favoured whiter flowers as quantified with reflectance spectrometry. This selection on flower colour was stronger when supplemental pollen was added, consistent with an effect of the trait on asymptotic seed production rather than an effect on pollen receipt.These results, along with previous behavioural studies of pollinators, show that pale flower colour, common in the habitat and often attributed to pollinator behaviour, is not maintained by pollen‐mediated selection but instead through another mechanism. This study illustrates the value of combining experimental pollen supplements and reductions with trait variation to study selection of floral traits, especially in species where compatible pollen receipt is difficult to measure.

Further Studies on Electrostatic Date Pollination – from the Laboratory Bench to Field Unit Performance Test

Artificial pollination of date (Phoenix dactylifera L.) blossoms is a common means of improving yield and quality. Date growers prefer the use of precise techniques that ensure minimal pollen consumption but require very large amounts of manual labor, however, they often use inaccurate aiming techniques to maintain their application schedule. An electrostatic field applicator was developed in the present work to reduce manual labor and pollen consumption in artificial date pollination. Labor saving was achieved by moving the system continuously, which requires a 2.5-m gap between the unit outlet and the target, in order to avoid collisions with branches. Such a gap is in contrast with efficient electrostatic applications of dry matter since it dilutes the pollen cloud and reduces the space charge. A need to increase the velocity of the pollen-carrying air stream to overcome the prevailing wind also reduces the space charge. Therefore, laboratory tests to define favorable parameters were essential to define the specific design and construction of an efficient unit. Based on the laboratory findings, an accurate-aiming electrostatic field applicator was designed, constructed, and mounted on a mobile elevated platform for use in date orchards. Tests with the field unit showed that it enabled savings of up to 90% and 70% in pollen consumption and labor use, respectively, compared with the commonly used units, while maintaining the high fruit set provided by the commercial devices. Electrostatic charging enabled reduction of pollen consumption to 0.125 g per tree in a single application. By applying excess pollen and reducing fruit set, the new system can save additional labor in the thinning operations.

Pistillate Flower Abcission Symptoms of ‘Serr’ Walnut (Juglans RegiaL.)

A B S T R A C T Pistillate flower abscission (PFA) severely affects walnut (Juglans regia L.) production, mainly of the cultivar ‘Serr’. A study on symptoms of flower abortion in walnut ‘Serr’, induced by excess or lack of pollen, was carried out following the evolution of flowers and peduncles. Hand pollinated flowers, with different pollen concentrations: 0, 1, 5, 25, 50 and 100% pollen diluted (w/w) in talcum powder, showed external and internal abortion symptoms. External symptoms were recorded directly from the tree, while for internal symptoms, flower excisions stained with blue toluidin-O at 0.05%, were analyzed under an optical microscope. Excess pollen abortion produced flowers with rapid damage. The growth rate was very low, involucre brightness was lost and the stigma turned necrotic, in contrast to the flower aborted from lack of pollen. Normal flowers continued growing to fruit harvest, without losing brightness. Excess pollen abortion induced flower drop, with or without attached peduncle. Two separation zones can be present: the distal and proximal areas of the peduncle. Peduncle scars allowed for differentiating abortion induced from excess or lack of pollen, or other reasons.

A theory for exaggerated secondary sexual traits in animal-pollinated plants

We analyze two mathematical models of adaptive investment in rewarding plant traits. In both models, the attractiveness of a particular trait value declines as the mean value in the population increases (asymmetric competition), giving relatively rewarding traits a competitive advantage. Including this competition for pollinator visits in a standard model of hermaphroditic sex allocation shifts additional allocation to pollinator rewards at the expense of allocation to pollen and seeds. In the second model, plants can invest additional resources in pollinator rewards but suffer reduced viability and rising costs due to excess pollen removal and within-plant selfing (geitonogamy). Despite these accumulating costs, increasing the magnitude of asymmetric competition exaggerates the ESS investment in rewards beyond the equilibrium in cases where attractiveness depends only on a plant’s absolute reward value. We suggest that the type of frequency dependent selection modeled here is fundamentally equivalent to sexual selection in animal populations (with some unique exceptions). Testing the main assumptions of our models may reveal whether seemingly “extravagant” floral traits are strictly analogous to the exaggerated secondary sexual traits of animals.

Benefits and costs of mutualism: demographic consequences in a pollinating seed–consumer interaction

Interspecific interactions can affect population dynamics and the evolution of species traits by altering demographic rates such as reproduction and survival. The influence of mutualism on population processes is thought to depend on both the benefits and costs of the interaction. However, few studies have explicitly quantified both benefits and costs in terms of demographic rates; furthermore there has been little consideration as to how benefits and costs depend on the demographic effects of factors extrinsic to the interaction. I studied how benefits (pollination) and costs (larval fruit consumption) of pollinating seed-consumers (senita moths) affect the reproduction of senita cacti and how these effects may rely on extrinsic water limitation for reproduction. Fruit initiation was not limited by moth pollination, but survival of initiated fruit increased when moth eggs were removed from flowers. Watered cacti produced more flowers and initiated more fruit from hand-pollinated flowers than did unwatered cacti, but fruit initiation remained low despite excess pollen. Even though water, pollination and larvae each affected a component of cactus reproduction, when all of these factors were included in a factorial experiment, pollination and water determined rates of reproduction. Counter-intuitively, larval fruit consumption had a negligible effect on cactus reproduction. By quantifying both benefits and costs of mutualism in terms of demographic rates, this study demonstrates that benefits and costs can be differentially influential to population processes and that interpretation of their influences can depend on demographic effects of factors extrinsic to the interaction.

Effect of Pollen Competition on Offspring Quality at Varying Stages of the Life Cycle in Silene latifolia Poiret (Caryophyllaceae)

from the ovary. We performed an experiment in which we placed excess pollen at one of two distances from the ovary. We then compared seed, seedling, and adult characters of the progeny resulting from these two treatments. Pollen placement had a statistically significant effect on adult sex ratio, with progeny resulting from more intense pollen competition containing a statistical excess of males, and progeny resulting from less intense competition containing a nonsignificant excess of females. Pollen placement did not have a statistically significant effect on any of seven measures of offspring quality (seed number per fruit, mean seed mass, seedling emergence time, cotyledon area after two weeks, leaf area after four weeks, above-ground dry weight after six weeks, and days required to flower). Since one of our dependent variables (seedling emergence time) was found to be significantly affected by pollen competition in an earlier study, it appears that such traits may be as much influenced by parental, environmental, or population-level effects as by pollen competition.

Reproductive lability in pre-dispersal biology ofRhamnus alaternus L. (Rhamnaceae)

Rhamnus alaternus L. is a dioecious, fleshy-fruited shrub, typical of Mediterranean vegetation, which has been recorded from the region since the tropical Tertiary. Reproductive biology of this species has been studied in Southern Italy over a period of 4 years with both field and laboratory experiments. In all the studied populations male plants predominate over the females, have a higher flower density and flower regularly every year while females are usually alternate-bearing. A strong negative association between fruit formation and vegetative growth has been found indicating that the presence of fruits on a branch reduces the growth of new shoots. This phenomenon, together with alternate-bearing, is evidence of a strong competition between vegetative growth and reproductive activity and indicates resource limitation. Flower morphology and quantity of pollen produced are typical of insect pollinated species, and flies of the Empididae and Muscidae were trapped bearing pollen at female plants but pollen is also dispersed by wind. Field experiments on pollination success showed that the exclusion of insects as pollen vectors produces a significant decrease in fruit development. Excess pollen provided by hand-pollination increases fruit set over open-pollination and reproductive success in the field is therefore pollen limited. The variability of sex expression, male fitness, fruit production, regrowth and pollination presented here suggests that these predispersal characteristics ofR. alaternus have been subject to evolutionary change since the tropical Tertiary climate was replaced by a Mediterranean one. This points to the possibility that reproductive, as well as morphological and physiological characteristics, may be equally selected for in extreme environments such as the Mediterranean.

Increasing Assimilate Reserves Does Not Prevent Kernel Abortion at Low Water Potential in Maize

Water deficits during anthesis cause maize (Zea mays L.) kernels to abort soon after fertilization. Inhibition of photosynthesis accounts for much of the kernel loss, but lack of assimilate reserves might also limit kernel development. We tested whether varying the level of C and N reserves prior to anthesis affected kernel set of plants pollinated at low water potential (Ψv). Plants were grown in sand/soil media in pots in the field under one of three light environments: Control, 75‐cm rows, 4.3 plants m−2; Shade, same as Control except under 55% shade cloth from V6 until maturity; or Isolated, 122‐cm rows, 0.7 plants m−2. These treatments produced a 3.3‐fold range in total extractable carbohydrate (TEC) and a 1.7‐fold range in total Kjeldahl nitrogen (TKN) per plant at silk emergence. Water was withheld at silk emergence and plants were pollinated by hand with excess pollen 4 d later when silk Ψw was = −1.1 MPa, leaf Ψw was = −1.8 MPa, and photosynthesis was completely inhibited. Plants were rewatered 2 d after pollination and remained well watered until physiological maturity. The brief water deficit decreased kernels per ear by 45% in the Controls, 72% in the Shade treatment, and 49% in the Isolated treatment. Lower kernel set in the Shade treatment was associated with low levels of TEC and TKN in the vegetative stalk, leaves, and reproductive stalk, compared with Control plants. However, accumulation of TEC and TKN in excess of Control values in the Isolated plants did not prevent kernel loss. Kernel set in all three treatments was more closely related to the rate of photosynthesis prior to imposing the water deficit than to the level of reserves accumulated prior to flowering. These results suggest that selection for high levels of reserves at pollination will not improve kernel set at low Ψw in maize.

High Pollen Loads Can Cause Pistillate Flower Abscission in Walnut

The role of pollen in abscission of pistillate flowers of Persian walnut (Juglans regia L.) cv. Serr was investigated over a 4-year period by controlled pollinations and pollen counts. Self-pollen, pollen from other walnut selections or cultivars, or dead pollen was applied at high and low doses to pistillate flowers enclosed in pollination bags. Unbagged, open-pollinated flowers and bagged, nonpollinated flowers served as controls. In all cases, presence of pollen significantly increased the probability of pistillate flower abscission (PFA). Dead pollen resulted in as much PFA as live pollen. Counts of pollen grains confirmed that PFA-type flowers had significantly more pollen than normal flowers. In the fourth year `Serr' pollen was applied to unbagged flowers of `Serr' and ten other Persian walnut cultivars, and the amount of PFA on the artificially pollinated flowers was significantly higher than on the open-pollinated flowers, while the control flowers dusted with talc or pine pollen had almost no PFA. These results clearly indicate that excess pollen is involved in pistillate flower abscission in `Serr' walnut and suggests that other cultivars may also be sensitive to pollen load. This phenomenon may have implications in the biology of selfing and evolution.

Variations in the number of beans per pod inTheobroma cacaoL. in the Ivory Coast. I. The role of pollen

Incomplete pod filling is frequently encountered in cocoa cultivation, but there is disagreement as to the causes. Pollen factors (amount and quality of pollen) were studied on several clones in the Ivory Coast. Hand pollination with compatible pollen showed that the pod filling of clone UPA 413 can be improved, but not that of clone SCA 6. The pollen genotype, and not just its compatibility, seems to affect the average number of beans per pod, but not the standard deviations. Experiments on the number of pollen grains deposited (pollination) looked at hand pollination (where there is always excess pollen), the life span of flowers depending on pollination conditions and the amount of pollen in fallen flowers. They showed that hand pollination with compatible pollen does not always guarantee complete pod filling, that many flowers pollinated with clusters of pollen can fall and that large amounts of pollen are required to obtain pods.

FLOWER VISITATION, POLLEN DEPOSITION, AND POLLEN‐TUBE COMPETITION IN HIBISCUS MOSCHEUTOS (MALVACEAE)

The potential influence of pollen‐tube competition on offspring “quality” has received considerable attention in recent years. Yet the prevalence of pollen competition in natural populations is largely unknown because few investigators have actually measured rates of pollen deposition on stigmas. In this study, we assess the potential for pollen‐tube competition in natural populations of the self‐compatible, pollinator‐dependent herbaceous perennial, Hibiscus moscheutos. Individual flowers averaged two to four visits per 15 min by potential pollinators (Ptilothrix and Bombus), and about 34% of these visits involved contact with a stigma. The median number of pollen grains deposited on virgin stigmas per contact visit was 70 grains (values ranged from 0 to 889), and flowers averaged about four contact visits per hour. Approximately 360 pollen grains must reach stigmas for full seed set to occur in a typical flower (an average ovary has 139 ovules, and 2.6 pollen grains are required per seed). Within 2 and 3 hr exposure to pollinators, 65% and 97% of the flowers received excess pollen (>360 grains) and median stigmatic pollen loads exceeded the number of ovules by a factor of 4.0 and 5.4, respectively. Based on 3 yr observations, it is concluded that pollen competition may frequently occur in this species.

Alternative Outcomes of Natural and Experimental High Pollen Loads

Seed production is usually assumed to be a positive monotonic function of pollen deposition and/or pollinator visitation. If this assumption were correct, there would be only two outcomes of excess pollen levels: an increase in fruit or seed set, or no increase. However, a substantial minority of the studies reviewed here has found that seed production declines with increased pollen loads, both under experimental and natural conditions. To explain this decrease, we propose the following mechanisms: pollen tube crowding, pollen removal or stigma damage by pollen thieves or pollinators, stigma damage during hand—pollination, application of low—diversity or local pollen, effects of bagging flowers, missed stigma receptivity, and the application of inviable pollen. These mechanisms can be distinguished through more complete and more careful experimental designs and incremental pollen supplementation.

INTERPOPULATIONAL VARIATION IN FRUIT PRODUCTION: THE ROLE OF POLLINATION‐LIMITATION IN THE OLYMPIC MOUNTAINS

Fruit set varied from a mean of 44% to 93% among six populations of Veronica cusickii L. (Scrophulariaceae) spanning a distance of 13 km in the Olympic Mountains, WA. Hand pollinations in three populations showed that this variation resulted primarily from differences in the level of limitation by insect pollination. In every pollination, hand‐pollinated flowers averaged at least 82% fruit set. In 1981 and 1982, addition of excess pollen significantly increased mean fruit set per plant in the Blue Mountain population, but not in two other populations with higher natural fruit sets. At Blue Mountain, hand pollination more than doubled the mean seed set of V. cusickii and also significantly increased seed set in two of three other species visited primarily by flies or solitary bees. Pollination‐limitation appears to be common at this site where the weather is frequently cold and damp.

Pollinator Sharing and Seed Set of Stellaria pubera: Competition for Pollination

I used field experiments to examine the effect of pollinator sharing on seed set in the understory herb Stellaria pubera. In piedmont North Carolina the solitary bee and fly pollinators of this species are also frequent visitors to Claytonia virginica. Removals of nearby flowers of C. virginica increased seed set of S. pubera. Examination of seed set in populations of potted plants indicated that the effect was due to change in species composition rather than plant density. Variation in seed production was high, however, and even when flowers were provided with excess pollen through supplemental hand pollinations, only 20% of ovules produced seeds. Competition for pollination appears to be one of several selective forces that act on blooming time of S. pubera. Stellaria pubera blooms slightly later than C. virginica. Whereas the seeds set of early flowers was increased 13% by removal of C. virginica and 19% by hand pollination, seed set of late flowers was unaffected. Plants forced to bloom early also tended to be more pollinator—limited than control plants. Removal of C. virginica did not, however, increase the seed set of S. pubera plants forced to bloom early significantly more than that of control plants. Other forces are probably also involved in the maintenance of blooming time.

Sexual Selection and Insect Sperm

Males produce the smaller gamete. The inconsequence of sperm in comparison to the enormity of eggs has dominated the evolution of male characteristics (Trivers 1972). Females, with their clutches of expensive ova, are limiting resources for which males compete in a variety of ways (Blum and Blum 1979). These different manifestations of masculinity have perhaps, in turn, influenced the radiation of insect sperm morphology and behavior. The following related questions arise from a consideration of sperm both as microorganisms and tools of male reproductive interests: i) Why are there more sperm than eggs? ii) Do sperm behave as individuals? iii) Do competition between ejaculates and conflicts of interest between females and the sperm they contain result in gametic adaptations? Notes on topics indirectly bearing on these problems are placed in an appendix. Why are sperm-to-egg ratios greater than 1? Some mites and insects transfer fewer than 2 sperm per ovum, but these are exceptions to the rule of tens or hundreds of sperm for every egg (Cohen 1971, 1975, 1977). Since only 1 of the multitude can genetically participate in the zygote, why go to the expense of producing such a mass? Males may swamp the female reproductive system with sperm to block the introduction of rival ejaculates (Parker 1970). A contrasting argument holds that gametic exuberance is largely symbolic, a display of a male's ability to obtain resources. Females may make decisions about which male's gametes to use based on the dimensions of their mates' ejaculates (Mary Willson 1979, discussing excess pollen per ovule). These hypotheses have 2 difficulties explaining the full range of insect sperm-to-egg ratios: i) Even in insects whose females mate only once, males transfer excess sperm. In the monogamous mosquito Aedes aegypti, for instance, the male passes 2000 sperm to a female who will lay about 85 eggs (Jones 1968, Christopher 1960). There are, in such cases, no rivals to block or postcopulatory choices between males to be made. ii) Neither specifies an advantage to females who accept excessive numbers of sperm. While the female sometimes expels or digests a portion of the male's ejaculate, the part she stores, and possibly maintains, is often still in excess of the number of ova. Of the 2000 Aedes sperm, about 1000 will make their way to the spermatheca (Jones 1968). Why, then, do males produce more sperm than a female keeps and why does she keep more than she needs? In some cases, males may pass large numbers of sperm as a nuptial gift, a contribution to the good health and fecundity of his offspring's mother (see Thornhill 1976, 1980). It is possible that many are digested in the female genital tract as a source of nourish-

A Model for the Evolution of Distyly

Reasons are given for believing that morphological distyly is unlikely to evolve before the incompatibility system. The first problem studied is therefore the evolution of incompatibility of the type found in distyly, with two pollen and two stigma types. It is argued that the first step, starting from self-compatibility, is a mutation to a new pollen type, such that the mutant cannot self-fertilize or fertilize other individuals. Conditions for the spread of such a mutation (assumed recessive) are given. Next, a dominant mutation to a new stigma type, incompatible with the original pollen type but compatible with the new type, is studied. Such a mutation is almost certain to be eliminated if it occurs first, but can spread in a population polymorphic for the pollen mutation provided that it occurs at a linked locus. With tight linkage, this model generates a population with two incompatibility types, one dominant to the other, with only rare recombinant types. It is shown by computer runs that modifiers causing tighter linkage will spread in such populations. The next problem studied is the evolution of differences in anther and stigma positions. These are assumed to reduce the selfing rate. It is argued that anther position is unlikely to change first, as this will reduce male fertility by as much as the reduction in the chance of pollen transfer to the stigmas of other flowers; a changed stigma position will probably affect female fertility less than this, assuming that excess pollen is initially present and that all ovules can be fertilized unless the amount of compatible pollen received falls below some threshold level. If the incompatibility system is perfect, a new stigma position can be selected for, but linkage to the incompatibility locus does not affect the results. If there is partial self-compatibility and the more self-compatible form is dominant, linkage is very important for the spread of dominant mutations affecting stigma or anther position; an unlinked mutation is frequently either eliminated or fixed under conditions in which a linked gene will establish polymorphism. Computer runs show that there is selection for tighter linkage between morphology mutations and the incompatibility locus, in situations where both are polymorphic. When a recessive mutation making the final morphological change is introduced, it spreads faster the tighter its linkage to the other loci. There is again strong selection for tighter linkage, and at equilibrium with tight linkage there are approximately equal frequencies of two phenotypes, one homozygous for all the recessive alleles, and one heterozygous at all the loci. The more self-compatible form is dominant and so usually is the form with the new stigma position. This agrees with many distyled species. Species in which the more self-compatible form is recessive cannot be explained by this model, but can be accounted for if the gene affecting the stigma reaction also controls its position. This case is also studied. It has properties similar to the other case; in particular, there is a strong linkage constraint for mutations with this type of effect when introduced into a population segregating for a pollen-reaction gene. This model can also explain species with differences in stigma position only, and with two equally frequent forms, each corresponding to one incompatibility type. Such populations are not easy to explain by the previous model.

Genetic Effects of Ultra-Violet Radiation in Maize

A high voltage mercury discharge tube of fused quartz emits radiation consisting chiefly of X 2537. A commercial tube of this sort (Hanovia (Sc-2537), operated at 5000 volts and 100 m. a., was used as a source of radiation in comparison with filtered radiations of the types described in the preceding paper. The discharge tube treatments were applied under conditions similar to those used with the mercury arc. Exposures were made at the same distance and with the same filter cell, with 1 cm. of distilled water in place of the filters used with the mercury arc. Under these conditions about 83% of the total intensity is X 2537. Other wave-lengths included are chiefly in the near ultra-violet and visible. Of the wave-lengths shorter than X 3130, more than 97% of the total intensity is X 2537. For the purposes of genetic experiments, therefore, this radiation is almost monochromatic X 2537. The ratio of intensity of the radiation from the discharge tube and the arc (the water filter being used in both cases) was for X 2537 approximately 1: 1, for total radiation shorter than X 3130 approximately 1:8. In experiments with the discharge tube, the multiple recessive seed parent used was A a C c wx wx su su pr pr y y. The pollen stock carried the dominant alleles of all of these genes. It was therefore possible to detect one-half of the losses of A and C (as indistinguishable color losses) and all of the losses of Wx, Su, Pr and Y. The maximum dose tolerated is considerably lower than the maximum tolerated dose of the filtered radiations reported above. Only the most approximate comparisons may be made in tolerance because of the variation with environmental conditions which has been mentioned. In treatments with the discharge tube, under favorable weather conditions and with excess pollen, well-filled ears are produced by the use of pollen irradiated for 30 seconds or less, but only partial seed setting occurs with pollen irradiated 60 seconds or more. A few seeds are produced by the use of pollen treated as long as 120 seconds. Under similar conditions, with filtered radiation A of the mercury arc (described in the preceding paper), well-filled ears are produced by the use of pollen irradiated 2 minutes and sometimes 4 minutes, and a few seeds are produced by pollen treated as