A non-dehiscent mutant (ndh1)
--Walden, DB; Cheng, PC

A non-dehiscent phenotype (Figures 1 and 2) was identified in Oh43 stock and has been observed in each of six generations following self-pollination. The pollinations have been made by collecting the anthers in a tassel bag, cutting/macerating the anthers and applying the resultant mix of pollen and cut anthers with the knife onto appropriate silks. Most such pollinations were done in the afternoon, long after pollen would normally have been shed. Seed set was sparse and scattered; several hundred non-dehiscent Oh43 plants have been observed and the trait has been observed in other inbreds after pollen transfer as described above. The trait can be transferred through the female. In all studies, segregation patterns predict a single recessive, nuclear allele for this trait. Mapping studies are not yet complete. Proposed symbol: ndh1. The expression of the non-dehiscent phenotype has been observed in the field (winter nursery in Hawaii and summer nursery in London, Ontario), in the glasshouse and in growth chamber (under both 12 hr/12 hr and 16 hr/8 hr light cycle, 25C condition) grown materials.

Specimens for microscopy were collected in our 1993 Hawaii winter nursery and from growth chamber grown material. The specimens were fixed in a tri-aldehyde fixative (Cheng, Walden and Greyson, Natl Sci Counc Monthly, ROC 10:1000-1007, 1979), postfixed in OsO4, dehydrated in acetone and critical point dried with CO2. Surface was sputtered with AuPd and examined with SEM.

In wild type material, at anthesis, the lodicules swell and pry apart the glume and lemma allowing the anthers to extrude. Soon after extrusion, pores at the tip of the anther break open and pollen is liberated. As described (Cheng, Greyson and Walden, Can J Bot 57:578-596, 1978), maize anther dehiscence consists of three major processes: (1) the swelling of lodicules to push the glume to open followed by the elongation of filaments; (2) separation of intermicrosporangial stripe 1 (IMS1) from the underlying parenchyma in the anther (Figure 3); and (3) the opening of an anther pore along the distal end of the IMS1 following the collapsing of epidermal cells. Figures 1 and 2 show that the filament elongation and opening of spikelets proceed normally in the homozygous mutant plant; however, the separation of intermicrosporangial stripe 1 (IMS1) from the underlying parenchyma of the anther connective never occurs; as a result, the anther pores fail to open (Figure 4). This work was supported by NSERC (DBW), the National Science Council of the Republic of China (NSC83-02110-B-00100-001)(PCC) and Academic Development Fund of SUNY (PCC).

Figure 1. Close-up view of center rachis showing non-dehiscing spikelets. Arrow indicates a shriveled filament which has elongated many hours earlier; however, the anther (double arrow head) has failed to open.

Figure 2. Image shows an incomplete anther dehiscent process caused by the failure of pore opening in the homozygous plant. The spikelets 1 and 2 are in the process of dehiscence. (Note the elongating filament); arrow indicates a shriveled filament which has elongated many hours earlier; however, the anther (double arrow head) has failed to open.

Figure 3. Cross-section of a wild-type anther showing the separation of IMS1 from underlying anther connective (co). This process does not occur in the homozygous.

Figure 4. Scanning electron microscopic image shows the distal portion of an anther from a homozygous plant. The anther was taken from a spikelet which had undergone anthesis for at least three hours; the IMS1 did not separate from the underlying connective tissue. The region indicated by the "*" represents the approximate range where the IMS1 should split open to form the anther pore. IMS2: intermicrosporangial stripe 2. 

Return to the MNL 71 On-Line Index
Return to the Maize Newsletter Index
Return to the MaizeGDB Homepage