The rice heterosis (hybrid vigor) has been known for a long time in rice production and research. Rice cultural practices vary across the state and US environmental and economic times  . Utilization of hybrid rice is an important technology to meet the increasing rice demand in world. Hybrid rice is more profitable, and more sustainable for yield production. Hybrid rice delivers about 15% - 20% per capita yield advantage over inbred rice and utilizing less water and pesticides to land    . The research of hybrid rice began in 1980s in USA and released first hybrid rice in 2000 by the Rice Tec company. Hybrid rice has been widely grown in the US now. It had covered about 40% of the rice acreage in Arkansas during 2012 to 2014     . The state represents about 50% of the total acres planted rice in USA.
Different germplasms from USDA (United States Department of Agriculture) world rice collection have been utilized and accessed in the hybrid breeding at UAPB since 2008. Some new sterile line, maintain lines, restorer lines and hybrids have been screened and bred from them. The identifying and evaluating activities have generated a lot of knowledge of hybrid breeding, selected resistant varieties and developed some hybrids from this research program  -  . Based on the research objectives, we had selected varieties that genotypes match up with phenotype from our previous study.
There is a period of very hot weather at the areas of Pine Bluff, Arkansas. There were 32 days with daily highest temperature over 34˚C in the 52 days from June 25-August 15 in 2012 (Table 1). The development of panicles would be greatly affected when the panicle differentiation (PD) of rice is under this condition. The heading stage of rice growth is about 25 days after the PD stage. Therefore, early PD and heading may be a way to reduce heat-induced sterility by avoiding the period of temperature above 34˚C. The best PD growing stage would be before June 25 and the best heading stage would be before July 20. It can avoid being hurt by high temperature in Arkansas. Therefore the earlier season hybrid rice varieties are also very important in the rice production. There are two ways to breed earlier season hybrids. One is using the normal sterile line which has no DEMG (Dominant Early Maturity Genes) to cross earlier heading restorer line to get earlier heading hybrids. Another way is using early maturity hybrids with DEMG (Dominant Early Maturity Genes) sterile female parent to cross with late heading restorer lines to get the earlier heading hybrids.
The later heading hybrid rice variety also can get the higher yield if their parents have heat resistant gene. They can head normally without be hurt by high temperature. Those later heading hybrid rice varieties were from none-DEMG sterile line mating late heading restorer lines.
Based on the research objectives, we selected varieties which genotypes match up with phenotype from our previous study.
2. Materials and Methods
32 hybrids were made from 4 sterile line 28s, 29s ,30s and 33A by separately crossing with 8 restorer lines PB-2R, PB-5R, PB-12R, PB-21R, PB-22R, PB-23R, PB-24 and PB25. 28s came from the F5 generation of the Gobo (PI-369806)/Zhenshan 97// Xiangzaoxian No. 1///Jin23, 29s came from the F6 generation of the E425 (PI-442935)//Le- mont/Zhenshan 97, 30s came from the F5 generation of the Madagascar 342 (PI- 317514)///Lemont/you-1//IR2061, and 33A came from the B5F1 generation of Ignap
Table 1. Maximum temperature of June to August 2012 in Pine Bluff, Arkansas.
Catelo (PI-373138)//II-32/Jin23. PB-2R came from the F5 generation of CDR22/ Katy/Minghui63, PB-5R came from the IR64//Katy/Guiyana 50781, PB-12 came from the Katy/Minghui63/Jasmine-85, PB-21R came from the Bengal/Minghui63/Jasmine-85, PB-22R came from the Katy/Minghui63//Gui 99, PB-23R came from the Katy/Ming- hui63//CDR210, PB-24 came from the Katy/Minghui63//Ce64, and PB-25R came from the Katy/Minghui63//IR24.
These 32 hybrids and their 8 male parents were tested for agronomic traits and yield at the farm of UAPB in 2012. Soil texture is silt loam with PH value of 5.3. The test entries were sowed at April 16 in the greenhouse and transplanted to field with 20 days old seedlings at 10 feet long, 1 foot space row, and 3 replications for each entry. Weeds were controlled with 9.3 L ha-1 of Propanil (3ʹ, 4ʹ-dichloropropionanilide) mixed with 0.4 kg∙ha−1 of quinclorac (3, 7-dichloroquinoline-8-carboxylic acid; Facet, BASF) when the rice were about four-leaf stage. Nitrogen fertilizer was applied pre-flood at 134 kg N ha−1 at about the five-leaf stage. The flood (underground water from a well) was maintained throughout the growing season. Heading dates were recorded when 50% of the plants were headed. Panicles were harvested about 40 days after heading. Plant heights were measured at harvest. Weights of 1000 grains and seed set rate were measured. Seed set rates were calculated by seeds per panicle divided by the total spikelet per panicle. Milled rice, head rice and yield were also measured. Daily maximum temperatures were recorded by the sensor of NRCS Arkansas scan sites which is 50 meters away from field of study  . Average yields and stand error of the mean were analyzed by SAS 9.2.
The results showed that yields of 27 hybrid rice were higher, 5 hybrids were lower and 8 restore lines were higher than check Francis (Table 2).
15 hybrids had yielded 20% higher than check Francis. The top 7 hybrids were 28s/PB23, 28s/PB24, 28 s/PB22, 33A/PB24, 30s/PB24, 33A/PB22, and 30s/PB23. They yielded 10846.6 kg/hectare, 10628.9 kg/hectare, 10549.8 kg/hectare, 10539.8 kg/hectare, 10490.4 kg/hectare, 10460.8 kg/hectare, and 10431.1 kg/hectare, and over CK 30.7%, 28.1%, 27.1%, 27%, 26.4%, 26%, 25.7%, respectively (Table 2).
The heading date of early maturity hybrids, 28s/PB23, 28s/PB24, and 28s/PB22, were July 13 or 14 and growing days of sowing to heading were 88 or 89 days (Table 2). These three hybrids were developed from sterile female parent 28s, which is with DEMG, by crossing with late restorer PB23, PB24, and PB22. They had higher yields than late maturity hybrids which developed from none-DEMG sterile female parent. These results indicate that their PD (the panicle differentiation stage) had avoided the high temperature period. To achieve higher yield under the similar weather condition, the PD should be before June 25 and heading stage should be before July 20. The growing stage of a variety from sowing to heading should be earlier than 90 days to achieve higher yield in the areas with high temperatures such as Pine Bluff, Arkansas.
The heading date of check Francis was July 15 and its growing stage of sowing to heading was 90 days. The hybrids 33A/PB24, 30s/PB24, 33A/PB22, 30s/PB23, 30s/PB22,
Table 2. Yield and heading date of Hybrid rice in UAPB 2012.
33A/PB25, 30s/PB25, 33A/PB12, 33A/PB2, and 30s/PB5 were late heading in from July 24-August 5 and their growing stage of sowing to heading were 111 days, 99 days, 107 days, 99 days, 99 days, 113 days, 100 days, 107 days, 108 days, and 97 days, respectively. They were 21 days, 9 days, 17 days, 9 days, 9 days, 13 days, 10 days, 17 days, 18 days, and 7 days later than check Francis, respectively. They headed in the highest temperature time but still had higher yields which were over check Francis 27%, 26.4%, 26%, 25.7/%, 22.2%, 21.4%, 20.7%, 20.7%, 20.5%, and 19.9%, respectively (Table 2). These 10 hybrids showed heat tolerant may indicate the existing of heat resistant genes.
Table 3 (horizontally) listed the average yields of hybrids from the same sterile line crossed with 8 different restorer lines, PB2, PB5, PB12, PB21, PB22, PB23, PB24, and PB 25. Hybrids developed from sterile line 30s had the highest average yield of 9763.9 kg/hectare and over CK 17.6%. Hybrids developed from sterile line 33A had the second high yield of 9743.0 kg/hectare and over CK 17.4%. Hybrids from sterile line 28s was the third high yield of 9700.9 kg/hectare and over CK 16.9%. The last was hybrids from sterile line 29s with an average yield of 9106.2 kg/hectare and over CK 9.7% (Table 3).
Table 3 (Vertically) listed the average yields of hybrids for the same restorer line crossed with 4 different sterile lines, 28s, 29s, 30s, and 33A. Hybrids from PB-24 crossed with the 4 sterile lines had the highest average yield 10369.2 kg/hectare and over CK 24.9%. It was also over male parent PB24 by 17.1%. Hybrids from PB22 crossed with the 4 sterile lines had the second high average yield of 10284.0 kg/hectare and was over CK by 23.9%. It was also over male parent PB22 by 22.2%. Hybrids from PB25 crossed with the 4 sterile lines had the third average yield of 9736.2 kg/hectare
Table 3. Yields of restore and the average yield of same restorer crossing with different sterile lines.
and was over CK by 17.3%. It was also over male parent PB25 by 12.7%. Hybrids from PB12 crossed with the 4 sterile lines had the forth average yield of 9726.3 kg/hectare and was over CK by 17.2%. It was also over male parent PB12 to 10.3%. Hybrids from PB23 crossed with the 4 sterile lines had the fifth average yield of 9721.4 kg/hectare and was over CK 17.1%. It was also over male parent PB23 by 10.4%. Hybrids from PB5 crossed with the 4 sterile lines had the sixth average yield was 9693.9 kg/hectare and was over CK 16.8%. It was also over male parent PB5 to 10.8%. Hybrids from PB2 crossed with the 4 sterile lines had the seventh average yield of 9563.1 kg/hectare and was over CK by 15.3%. It was also over male parent PB2 to 12.1%. Hybrids from PB21 crossed with the 4 sterile lines had the lowest average yield of 7534.0 kg/hectare and was lower CK by 9.2%. It was also 9.9% lower than male parent PB21 (Table 3).
Plant heights of 32 new hybrid varieties were 105 cm - 126 cm, and their 8 restorer parents were 99 cm - 118 cm. The most hybrids were taller than the male parents. Seed set of most hybrids were 79.3% - 91.5% except the hybrids from PB21 crossed with 4 sterile lines. The seed set rate of hybrids developed from PB21 were 64.6% - 75.2% and the seed set rate of their parent PB21 was 74.8%. The milling rice of 32 new hybrids were 68.2% - 72.5% and 70.9% for check Francis. Head rice of 32 new hybrids were 48.4% - 61.7% and lower than Check Francis (62.6%). The grain weight of the 32 new hybrids were 26.6 - 32.4 g/1000 grain and over check Francis (21.3 g/1000grain) by 24.9% - 52.1% (Table 4).
28s is a very good sterile line with DEMG and has good combining ablating trait with most restorers. It can be used in making early heading hybrids which can avoid the high temperature period and obtain a high yield. Heading days of its hybrids should be 86 - 89 days and about 20 days earlier than male restorer. 30s is very good sterile line with none-DEMG and has good combining ablating trait with most restorers. It can be used in making later heading hybrids which should be resistant to high temperature. The heading days of its hybrids were 97 - 100 days and 10 days earlier than male restorer. 33A is also very good sterile line with none-DEMG which can be used in making later heading hybrids which should be resistant to high temperature. The heading days of its hybrids were 108 - 113 days which as same as male restorer.
PB24 and PB22 are good restorers and have good combining ablating trait with all 4 sterile lines.
Hybrid 30s/PB5 is a short plant height hybrid (105) cm and can be used in higher fertilizer condition. The yield of 5 new hybrids 28s/12, 28s/23, 29s/PB22, 30s/PB23, and 33A/PB22 were over check Francis by 20% - 36% respectively. But their plant heights were tall (120 - 126 cm) and could be lodging. They could better be used under the low fertility condition such as organic rice field. Most of the hybrids were about 110 cm and they can adapt in not high fertility condition.
The head rice rates of most hybrids were lower. But there were still some good hybrids with higher head rice rate. The head rice rate of 28s/PB-2 was %59.9%, the head
Table 4. Some other treats of hybrids and male parents.
rice rate of 28s/PB12 was 61%, the head rice rate of 28s/PB 21 was 59.8%, the head rice rate of 28s/PB24 was 61.7%, the head rice rate of 28s/PB25 was 59.5%, and the head rice rate of 29s/PB12 was 59.8%, the head rice rate of 30s/PB 12 was 59.4%. 28s was a good sterile line for hybrid rice with higher head rice rate. PB12 was a good restore line for hybrid rice with higher head rice rate. They can be used selectively in the breeding and the seed production.
 USDA, National Agricultural Statistics Service. Crop Production.
 Yan, Z., Yan, W.G., Deren, C.W. and McClung, A. (2011) Hybrid Rice Breeding. B.R. Wells Rice Research Studies 2010. University of Arkansas Agricultural Experiment Station Research Series 591, 61-63.
 Virmani, S.S., Sun, Z.X., Mou, T.M., Jauhar, A. and Mao, C.X. (2003) Two-Line Hybrid Rice Breeding Manual. IRRI, Los Banos.
 Hardke, J.T. (2014) Arkansas Harvested Rice Acreage Summary Preliminary Report.
 Huang, B.H., Yan, Z.B., Yan, W.G. and Deren, C. (2015) Changes of Sterility to Some Two Line System Male Sterilers in Different Rice Growing Seasons at Pine Bluff, AR USA. Southwest China Journal of Agricultural Sciences.
 Huang, B. and Yan, Z. (2015) Utilizing Dominant Early Maturity Genes of Sterile Line UP-3s in Hybrid Rice Breeding to Avoid High Temperature Season. American Journal of Plant Sciences, 6, 2596-2602.
 Huang, B., Ntamatungiro, S., Yan, Z. and Yan, W.G. (2010) Genotypic Differences of Straighthead to Rice Evaluated in Natural Conditions. Arkansas Environmental, Agricultural and Consumer Sciences Journal, 10, 11-16.
 Huang, B. and Yan, Z. (2016) Straighthead and Agronomy Characters Testing for Selected Parent Lines and Hybrid Combinations at UAPB in 2011-2012. American Journal of Plant Sciences, 7, 1423-1428.
 NRCS Arkansas Scan Site UAPB.