Search

The origin of polystichy in the maize ear and central tassel spike continues to challenge our understanding of evolution in this important crop species. In this paper we tested the hypothesis that the change in phyllotaxy occurs in the region of husk leaf production before the transition to reproductive growth. Young ear or presumptive ear primordia were dissected to examine the transition from distichous husk leaves below the ear through spiral phyllotaxy to the polystichous arrangement of spikelet pair primordia in the young ear. Serial transverse sections were used to document the thickness of successive disks of insertion of lateral primordia and to reconstruct the path of procambial differentiation. The transition in phyllotaxy, though variable, typically occurs in the vegetative zone and is associated with periodic heterogeneity in the thickness of leaf bases and a delay in the development of waves of procambial differentiation into the base of the young ear.

The tassel seed mutations ts4 and Ts6 of maize cause irregular branching in its inflorescences, tassels, and ears, in addition to feminization of the tassel due to the failure to abort pistils. A comparison of the development of mutant and wild-type tassels and ears using scanning electron microscopy reveals that at least four reproductive meristem types can be identified in maize: the inflorescence meristem, the spikelet pair meristem, the spikelet meristem, and the floret meristem. ts4 and Ts6 mutations affect the fate of specific reproductive meristems in both tassels and ears. ts4 mutants fail to form spikelet meristems from spikelet pair meristems. Ts6 mutants are delayed in the conversion of certain spikelet meristems into floret meristems. Once floret meristems are established in both of these mutants, they form florets that appear normal but fail to undergo pistil abortion in the tassel. The abnormal branching associated with each mutant is suppressed at the base of ears, permitting the formation of normal, fertile spikelets. The classification of the different types of reproductive meristems will be useful in interpretation of gene expression patterns in maize. It also provides a framework for understanding meristem functions that can be varied to diversify inflorescence architectures in the Gramineae.

The ontogeny of staminate tassels and pistillate ears in the maize mutant Fascicled ear was examined using scanning electron microscopy. The normal pattern of inflorescence development is perturbed by the Fascicled ear mutation at the transition stage. The Fascicled ear mutation promotes the development of an abnormal transition stage axis that is both shorter and broader than the wild type. The inflorescence apical meristem then undergoes a bifurcation, and two inflorescence axes arise in place of a single axis. Each derived inflorescence apical meristem may undergo a similar perturbation sequence. This expression of the Fascicled ear mutation may be repeated one to several times, which leads to the development of a fascicled pistillate inflorescence and a fascicled central spike in the staminate inflorescence. The apical meristems of some tassel branches are also bifurcated. Subsequent organogenesis during paired-spikelet and floral development in Fascicled ear plants follows the pattern of normal maize. However, triplet spikelets are occasionally observed. The organogenic disruption by the Fascicled ear mutation that we describe will aid genetic and molecular analysis on the regulation of inflorescence development in maize and other members of the genus Zea.

Tassel and ear primordia were collected from greenhouse-grown specimens of the Mexican maize landrace Chapalote and prepared for scanning electron microscopic (SEM) examination. Measurements of inflorescence apices and spikelet pair primordia (spp) were made from SEM micrographs. Correlation of inflorescence apex diameter with number of spikelet ranks showed no significant difference between tassels and ears, except at the two-rank level where the ear apical meristem had a significantly smaller diameter than corresponding two-ranked tassels. Within individual inflorescences, spp in different ranks enlarged at comparable rates, although the rates from one ear to the next along the stem differed. In both tassels and ears, spp divide to form paired sessile and pedicellate spikelet primordia when the spp is 150 [xm wide; ear axes are significantly thicker than tassel axes at the time of bifurcation. The similarities in growth between ear and tassel primordia lend further support to the hypothesis that both the maize tassel and ear are derived from a common inflorescence pattern, a pattern shared with teosinte. Inflorescence primordial growth also suggests that a key character difference between teosinte and maize, distichous vs. polystichous arrangement of spikelets, may be related to size of the apical dome and/or rate of primordium production by the apical meristem. There appears to be more than a single morphological event in the shift from vegetative to reproductive growth. The evocation of axillary buds (ears) is independent of, and temporally separated from, the transition to flowering at the primary shoot apex (tassel).

barren stalk1 is a recessive mutant of maize that has no tassel branches, spikelets, tillers, or ears. Here we present a detailed characterization of the ba1 mutant phenotype, including scanning electron microscopy of developing inflorescences, in situ hybridization analysis using a meristem marker, molecular mapping, and genetic analysis demonstrating an epistatic relationship between ba1 and teosinte branched1 (tb1). These data show that the primary defect in the ba1 mutant is a failure in axillary meristem development.

Some have postulated that highland Mexican maize was derived from an ancient high-altitude teosinte and that later introgression between the two taxa occurred. We used scanning electron microscopy to examine the inflorescence development in both the tassel and ear of a high-altitude Toluca teosinte. One of the most interesting observations was the presence of atypical multiranked orthostiches in the central spike of some male Toluca teosinte inflorescences. Most tassels exhibited a central spike with a pure, four-ranked, tetrastichous phyllotaxy or an intermediate (distichous/tetrastichous) phyllotaxy. A few A 1 tassels had a more typical distichous (two-ranked) central spike. Most ears showed the two-rank condition expected for teosintes. However, three ears displayed an intermediate (distichous/tristichous or distichous/ tetrastichous) phyllotaxy and one ear was tetrastichous. Our analysis of spikelet and floret development in all Toluca inflorescences revealed a pattern similar to that in landrace and U.S. maize, as well as to their close relatives, the teosintes. We suggest that this investigation may reveal inflorescence development in a natural maize-teosinte hybrid. This study further supports our hypothesis that both maleness and femaleness in the Zea inflorescences are derived from a common developmental pathway and underpins a proposal that andropogonoid grasses share a common pattern of inflorescence development.

The normal pattern of maize floral development of staminate florets on the terminal inflorescence (tassel) and pistillate florets on the lateral inflorescences (ears) is disrupted by the recessive mutation tassel seed 2. Tassel seed 2 mutant plants develop pistillate florets instead of staminate florets in the tassel. In addition, the ears of tassel seed 2 plants display irregular rowing of kernels due to the development of the normally suppressed lower floret of each spikelet. The morphology of tassel and ear florets of the recessive maize mutant tassel seed2 has been compared to those of wild-type maize through development. We have identified the earliest stages at which morphological signs of sex differentiation are evident. We find that sex determination occurs during the same stage on tassel and ear development. Early postsex determination morphology of florets in wild-type ears and in tassel seed 2 tassels and ears is identical.

Inflorescences of Tripsacum dactyloides (L.) L. (Andropogoneae) are characterized by single female spikelets at one to several basal nodes and paired male spikelets at several nodes above them on each raceme. Female spikelets are one-flowered and male spikelets are two-flowered. A sex form variant was found in a wild population in north central Kansas and classified as T. dactyloides (L.) L. forma prolificum Dayton et Dewald. The variant of this native distant relative of maize (Zea mays L. spp. mays) differs from the normal form by having both pistillate and perfect rather than staminate spikelets in the terminal (tassel) portion of the inflorescence and by having two functional pistillate florets in the basal spikelets instead of one. A recessive major gene at a single locus regulates the change of the inflorescence from monoecious to gynomonoecious. TRIPSACUM DACTYLOIDES (L.) L. (Andropogoneae) is a perennial, warm season bunchgrass which is widely distributed from Connecticut to Kansas and south to Paraguay and adjacent Brazil (de Wet, Harlan and Brink, 1982). Its inflorescences consist of one to several subdigitately arranged racemes. The grass is monoecious with the apical two-thirds to threefourths ofthe racemes composed of male spikelets with one to several female spikelets in the basal portion of racemes which are individually sunken into indurated rachis internodes. The female spikelets are solitary with one functional floret; whereas, male spikelets are born in pairs with two functional florets and the glumes are more or less coriaceous. A variant sex form of T. dactyloides with a gynomonoecious breeding system was found in 1981. The variant has been classified as Tripsacum dactyloides (L.) L. forma prolificum Dayton et Dewald (Dewald and Dayton, 1985). This paper describes the morphology and in

The origin of morphological differences attributable to mutant genes c2.n be identified at certain switch points in the developmental pathway. The effect of the Polytypic (Pt) gene is a stimulus to meristems in the developing inflorescence to continuous initiation of differentiating structures. The most severe expression is a suppression of meristematic activity. The action of the gene is a superimposition of its effect on the normal developmental pathway. The ramosa-1 (rat) gene interrupts the normal sequence of events, at the switch point, which would normally result in spikelet formation, and there is produced instead a lateral branch. The response of numerous switch points in the developmental pathway of the maize inflorescence supports the conclusion that meristems are a plastic system genetically programmed at successive intervals. THE GENESIS of form is one of the foremost problems in biology, but its solution is evasive. It is necessary to study development at all levels to elucidate the factors underlying morphogenesis. Investigation at the organismal level is a challenging and necessary aspect of the study. An organism matures through a series of sequential steps ordered by the genetic complement. An interruption of these events resulting in a change in the mature form permits investigation of the effect of specific factors. The succession of stages in development can be diverted by hormones, surgery, control of nutrition, and by light and temperature. Each of these has provided information useful in understanding morphogenesis. The approach that is the basis for our study is to investigate divergent development resulting from the substitution of a mutant gene for its normal counterpart. This enables one to study plants which vary from the control in a minimal way. The effect of the gene should be traceable to a single biochemical lesion, and thus the investigation is not confused by treatments with chemicals and other devices. Studies of mutant lines can indicate the points in development at which normal alleles at certain loci supply additional information leading to normal development. Our studies with mutants of maize affecting ear and tassel development lead to the conclusion that these "switch points" are numerous. There is no point at which a meristem can be considered to be irrevocably determined for a particular course of development. The meristem is, rather, a plastic system which must be programmed at successive intervals. 1 Received for publication April 9, 1963. This work was supported in part by the National Science Foundation (Grant G-16036). The authors gratefully acknowledge the assistance of Evart Cable in the photography and microtechnique. Journal Paper No. 2085 of the Purdue University Agricultural Experiment Station. Zea mays L. provides material for this kind of investigation. Tassel and ear mutants are abundant, and many have been carefully studied genetically. The sequence of events which gives rise to the normal ear of corn is relatively uniform and predictable. The development of the normal ear and tassel has been described by numerous investigators (Weatherwax, 1917, 1935; Collins, 1919; Arber, 1934; Bonnett, 1940; Anderson, 1944; Mangelsdorf, 1945; Cutler and Cutler, 1948; Reeves, 1946, 1950; Kiesselbach, 1949; Nickerson, 1954; Heslop-Harrison, 1961). One can trace the developmental pathways in the maize inflorescence by noting changes in meristematic activity beginning with the vegetative meristem and ending with the initiation of the ovule. Potential switch points at which diversion from the normal pathway may occur are illustrated in Fig. 1. The vegetative meristem gives rise to a leaf primordium by forming a subterminal ridge of meristematic tissue perpendicular to the longitudinal axis (Fig. 1A). The switch to an ear meristem involves a change in size and shape of the axis meristem and the initiation of longitudinal rows of branch primordia in an acropetal sequence (Fig. IB, 7). The branch meristem produces 2 spikelet primordia (Fig. IC, 7), each giving rise to 2 glumes ID, 7). Upper and lower floret meristems are then derived from the spikelet meristem (Fig. 1E). The initiation of the palea, lemma, lodicules and stamens (Fig. IF, G) by the floret meristems varies in extent and sequence. The floret meristem may then give rise to the pistil primordium (Fig. 1H), and, finally, to the ovule primordium (Fig. 1I). The ultimate nature of the structure produced by this series of events is dependent on the suppression, stimulation, repetition and order of occurrence of each step. Heslop-Harrison (1961) showed that photoperiod, temperature, cr auxin will affect a switch in the developmental pathway

NICKERSON, NORTON H. (Washington U., St. Louis, Missouri.) Sustained treatment with gibberellic acid of maize plants carrying one of the dominant genes Teopod and Corn-grass. Amer. Jour. Bot. 47(10): 809-815. Illus. 1960.-Groups of field-grown plants of 2 dominant maize mutants, Corn-grass (Cg) and Teopod (Tp), were treated with either distilled water or with 1 of 3 concentrations of aqueous gibberellic acid (GA) every 3 days from the seedling stage until tassel emergence. Both dominant mutants were found to respond to GA in such manner that certain treated plants became essentially normal in phenotype. The role of GA in modifying expression of specific genes is briefly discussed. EFFECTS of gibberellic acid on different plants are being reported with increasing frequency. Stowe and Yamaki (1957) presented the first complete review; recently (1959) they have made a short but up-to-date survey. Brian (1959) has covered recent literature more thoroughly and has made an attempt to integrate some of the seemingly controversial effects reported. The first instance in which plants of known genotype were phenotypically modified by external application of gibberellic acid (hereinafter called GA) was reported by Phinney (1956) in his work with maize. Brian and Hemming (1955) also obtained a similar reaction to GA by dwarf peas, but the peas they employed were a commercial variety of unilknown genotype. Other work by Plummer and Tomes (1957) on dwarf tomatoes, by Cooper (1958) oni dwarf perennial rye-grass (Lolium), and by Irvinie and Freyre (1960) on dwarf Tephrosia has shown that gibberellic acid could overcome effects of these specific recessive genes such that treated plants in each case assuLmed proportions of normal plants. All the instances cited involved recessive genes; this paper deals with effects of GA on 2 dominant genes found in maize, Corn-grass (Cg) and Teopod (Tp). Both these genes may be reduced in expression by appropriate treatment with GA to the point where plants that carry them are essentially normal in appearance. MATERIALS AND METHODS.-A limited supply of seed2 of the 2 stocks containing these genes was planted on June 20, 1958. Both samples were derived from open-pollinated ears, but plants were in such positions in the field that pollen available was only from siblings. Each of the seed stocks was shelled from a single ear which exhibited the genetic clharacter being investigated. Germination 'Received for publication MIay 18, 1959. This work was part of a research program carried out during the summer of 1958 while the author held a National Science Foundation Science Faculty Fellowship. Thanks are extended to Dr. B. 0. Phinney who kindly consented to review this manuscript. 2 Seed stocks were kindly supplied by Dr. William L. Browni, Pioneer Hi-Bred Corn Company, Johnston, Iowa. was nearly 100%. Each of the 2 stands of seedlings was divided into 4 equal groups. Each Cg group contained 9 plants; each Tp group contained 17 plants. On July 9, about 2 wk. after germination, young plants had 3-4 leaves, and treatments were begun. Each third day thereafter, until tassel emergenice, 1 ml. of distilled water only for control plants and for others 1 ml. of distilled water containing a particular concentration of GA 3 was administered to each of the respective apical leaf cavities. A total of 14 treatments was administered before tassel emergence. In cases where excessive tillers developed, only the original main leaf cavity was treated. Treatments are summarized in table 1. The groups were not randomized, but another experiment in the same field (Nickerson, 1959) showed conclusively that results obtained with identical concentrations of GA were independent of field location for any particular strain of maize. GA solutions wvere freshly made each week and kept in darkness at 19?C. To eliminate any possible effect of interaction between insecticides, fungicides and GA. no snraving was done either for control of fungi, of which none was noted, or for insects. of which both corn earworm and corn borers were noted. Fertilizer (Agrico 5-10-10) was applied at the rate of 600 lb. per acre during the fourth week after planting. RESULTS. Plants containing the dominant Corngrass gene were found to respond to treatment with GA in direct relation to dosage. The numbers of Cg plants which occurred in each treatment group of 9 plants were 5. 5, 5 and 4 respectively. Tke genotypes of the 'lants are thus assumed to be either +/+ or Cg/+, the combinations to be expected if an ear on a Cg/+ plant were pollinated with pollen produced on a +/+ plant. The results noted below are based on a coromarison of all mutant plants in each group. Each Cg-carrying plant could be easily identified at the third leaf stage by its narrow, glaucous blue-green leaves. 'GA used in this experiment was supplied by courtesy of Dr. Curt Leben, Agricultural Research Division, Eli Lilly & Company, Greenfield, Indiana. rThe JOURNAL for November (47: 713-808) was issued November 14, 1960.1 AMERICAN JOURNAL OF BOTANY, Vol. 47, No. 10. December, 1960

Comparison was made of 10 monoploid and 10 diploid plants in each of 6 lines (3 inbred lines and 3 monoploid-derived homozygous diploid lines) of maize to test the hypothesis that the monoploid, having, presumably, half the chromosomal volume per nucleus of the related diploid, would be proportionially scaled down in all dimensions and, having a complete gene complement, would retain the same number of parts as the diploid. In fact the monoploids were found to be slightly smaller than expected, with, in general, fewer parts than their diploid sibs. The difference was of the order of 11% in terms of both number of parts and linear dimensions, with area and volume measures proportionately scaled. It is suggested that the deviation from expectation could have been due to precocious nuclear and cell division in the monoploid consequent to the relation of chromosomal and nuclear volume to cell surface area, or the bias could be attributed to an environmental disadvantage peculiar to the monoploid plants. MONOPLOID sporophytes of maize (Zea mays L.) are smaller than genotypically equivalent homozygous diploids; otherwise, they appear to be similar to their diploid sibs (Fig. 1), being vigorous if the homozygous diploids are vigorous and weak if the diploids are weak. The question which prompted this study is whether the difference in size involves simply a scaling-down of all dimensions, proportionate to the volume of chromosomal material per nucleus of the monoploid (this being, presumably, one complete half that of the diploid), without change in numbers of parts, or whether non-proportionate changes in dimensions and changes in numbers of parts are also involved. Randolph, Abbe, and Einset (1944) reported a comparison of shoot apex and leaf development, and structure, in heterozygous diploid and tetraploid maize. They found that the shoot apex "... during plastochron 10 had the same number and arrangement of cells' in diploid and tetraploid, '. . . but that the volume of the nuclei and cells was doubled." "At maturity there was a marked tendency. . . for the tetraploid leaf to be approximately the same length as the diploid, about 1.16 times as wide, and 1.35 times as thick. The volume of the tetraploid leaf at maturity was thus about 1.6 times the volume of the mature diploid leaf." These authors developed the concept of the ideal gigas state, based on the relation of chromosome numbers in the diploid and tetraploid. This ideal state ". . . is attained when the volume of the various structures in the tetraploid including cells, tissues, and organs, is doubled that of the diploid, but the relative dimensions of these structures remain the same; that is, when the external form of the tetraploid structural unit is the same as that of the diploid but its volume is doubled... I Received for publication February 2, 1964. Randolph et al. found that the ideal gigas state was attained in the shoot apex of the tetraploid maize, but was modified in the mature leaf. If the physical relations among genotypically equivalent monoploids, diploids and tetraploids were based solely on the difference in volume of chromosomal material, the following ratios would obtain: Monoploid Diploid Tetraploid Chromosome numbers 10 20 40 Attributes measured: Number of plant parts 1.00 1.00 1.00 Linear dimensions 0. 79 1 .00 1 .26 Areas 0.71 1.00 1.41 Volumes (or weights) 0.50 1.00 2.00 MATERIALS AND MEASUREMENTS-To study these relations in homozygous diploids and their related monoploids, 10 monoploids and 10 diploids from each of 6 lines were examined. Three inbred lines (W153R, ND203, and DeKalb 0014) and 3 monoploid-derived hoimozygous diploid lines (DeKalb 7016, DeKalb 7010, DeKalb 7071) were used (for the method of development of the monoploid-derived lines see Chase, 1947, 1949a,b,c,d, 1951, 1952a,b). The sample plants were taken from increase plots at time of silking. An attempt was made to select freestanding plants in order to minimize the effects of competition. The measurements presented here are the averages of the 10 plants of each subset. In this n :2n comparison, the attributes studied were: (1) position of ear, counting nodes down from tassel; (2) number of tillers; (3) number of tassel branches plus leader; (4) number of nodes, counting each one, including those below ground level; (5) number of husks on ear, including prophyll; (6) number of rows of spikelets (kernel row number) per ear; (7) number of spikelets per ear

THE FEMALE INFLORESCENCE, or ear, of Zea mays L. has been the subject of much research, but there has never yet been offered a wholly acceptable explanation of its morphology. In order to provide a sound basis for a study of variation in cobs of Indian corn or maize (Nickerson, 1953), the problem has been reinvestigated. Whether or not the interpretation here advanced is correct, the facts themselves must be considered in any explanation of maize ear morphology. In their classic work, Mangelsdorf and Reeves (1939) summarized botanical opinion on maizeear morphology up to that time. They pointed out that maize had a perfect-flowered ancestry, and that monoecism has developed by abortion of pistillate organs in the upper inflorescences and of staminate organs in the lower ones. They also emphasized the essential homology of ear and tassel, a fact already recognized by Collins (1912) and Weatherwax (1918), and further strengthened by the work of Anderson (1944), Anderson and Brown (1948), Laubengayer (1949), Reeves (1950, 1953), and Bonnett (1953). Mangelsdorf and Reeves (1939), Reeves (1950, 1953), and Reeves and Stansel (1940), agreed with Weatherwax (1935, 1950) that the origin of both ear and central spike results from reduction of branches of a panicle to one pair of spikelets for each member. Bonnett (1940, 1948, 1953) showed that lateral spikelet initials were homologous with branch initials in maize, and pointed out that in the ear they are subtended by ridges similar to leaf initials in barley, oats and wheat. Stephens (1948) stated that paired spikelets of maize may be regarded as greatly reduced branches, and showed that this reduction is not peculiar to either tassel or ear but is a cumulation of processes initiated in the vegetative phase. The conclusion which can be drawn from these investigations is that the maize ear arose phylogenetically by a telescoping and reducing of parts already present in the maize progenitor. This concept of suppression, condensation, telescoping or aggregation can be identified in some of its minor aspects throughout the maize plant, and extension of it to previously unconsidered units provides the key to understanding much of the heretofore puzzling morphology of the ear. Anderson (1944) noted that condensation is

THIS INVESTIGATION concerns the pollen aberrancies in F1 maize plants from crosses of untreated female plants by tassels exposed in a nuclear reactor, and the pollen and chromosomal aberrancies in the progenies from outcrosses of these F1 plants as males to normal plants. In addition, data are given for the distribution of the transmitted chromosomal breaks among chromosomes, their arms and section of arms. Various workers have reported studies with maize on the transmitted aberrations induced by X-rays and by the atomic bomb radiations at Bikini. Using tassel sectors of abnormal pollen as an index of structural changes in the chromosomes, Anderson (1948) crossed plants from irradiated seeds with those from untreated or lightly treated seeds. When the induced changes were transmitted through the male, the frequencies of partial-sterile plants in the progenies were 6.0 per cent for 15,000 r. of X-rays and 4.7 per cent for the atomic bomb experiment. Transmission through the female gave 7.8 per cent partial-steriles for X-rays and 6.4 per cent for the atomic bomb. Anderson stated that thiese frequencies might be slightly higher than expected, since there was some selection in the direction of partial sterility. Later studies (Anderson in private communication) gave values of less than 4.7 per cent partial-steriles for both males and females. In the same study Randolph et al. (1948) selected plants from irradiated seeds on the basis of visible chromosomal abnormalities and crossed them with untreated or lightly treated plants. In the resulting progenies the frequencies of plants with abnormal pollen were 7.3 per cent for the X-ray treatment and 6.9 per cent for the atomic bomb exposure. In both series it was found that 50 per cent abnormal pollen was the modal class, although there were a few deviations each way. The distribution of breaks among and within the various chromosomes has been observed in a number of plant genera and in Drosophila. Longley (1950), working with progenies of maize seeds which had been exposed to the atomic bomb at Bikini and to X-rays, made a detailed study of 1176 chromosomal breaks at the pachytene stage

LEA, D. E. 1947. Actions of radiations on living cells. New York. SCHMIrr, F. O., AND B. UHLEMEYER. 1930. The mechanism of the lethal effects of ultrasonic radiation. Proc. Soc. Exp. Biol. and Med. .'7,: 626-628. WALLACE, R. H., R. J. BUSHNELL, AND E. H. NEWCOMER. 1948. The induction of cytogenetic variations by ultrasonic waves. Science 107: 577-578. WooD, R. W. 1939. Supersonics. The science of inaudible sounds. Brown Univ. Press. , AND A. L. LOOMIs. 1927. The physical and biological effects of high frequency sound waves of great intensity. Phil. Mag. and Jour. 4:417-436. YAMAHA, G., AND R. UEDA. 1939. Ueber den Einfluss der Ultraschallwellen auf die Wurzelspitzenzellen von Vicia faba L. Bot. and Zool. 7:41001-1007.

The ontogeny of tassels and ears in two annual Mexican teosintes, Zea mays subsp. mexicana and Z. mays subsp. parviglumis, was examined using scanning electron microscopy and light microscopy. Ear development in these annual teosintes follows a pattern previously described as leading to the bisexual mixed inflorescence in Z. diploperennis. Common bud primordia are initiated in the axils of distichously arranged bracts along the ear axis. These common primordia bifurcate to form paired sessile and pedicellate spikelet primordia. Development of pedicellate spikelets is arrested leaving the sessile spikelets, along with the adjoining rachis segment, to form solitary grains enclosed within cupulate fruitcases. Development of the central tassel spike is similar to that previously described in the Z. diploperennis tassel, except that the first formed axillary bud primordia form precocious tassel branches. The origin of these tassel branches suggests a possible mechanism for the transition from a distichous spike, characteristic of teosinte, to a polystichous spike, typical of maize.