Plant Physiol. 1966 January; 41(1): 111-114.
Inheritance of Factors Affecting Floral Primordia Initiation in Cestrum; Hybrids of C. elegans and C. nocturnum
Wesley O. Griesel
Department of Botany, California State College at Los Angeles

Summary. Photoperiod patterns of hybrids of Cestrum elegans (Brongn.) Schlect., a day neutral plant, and C. nocturnum L., a long-short day and long day plant, were investigated. Plants of the F1 generation, F2 generation, and backcrosses to each parent were tested on short day, long day, continuous light, long-short day and short-long day for floral primordia initiation. The data recorded suggest 2 independent genes or gene groups controlling floral primordia initiation in C. nocturnum, a single dominant gene that is activated by long-short day treatment and a recessive gene or genes responding to long day treatment. Further, these data suggest that the day neutral condition in C. elegans is the result of the series of independent genes or gene groups that respond to various photoperiods, the combination of these genes resulting in floral primordia initiation on all photoperiods.

The inheritance of factors affecting floral primordia initiation is a fundamental aspect of this field of study in which data are lacking. This paper presents exploratory data concerning the inheritance of such factors in Cestrum elegans and Cestrum nocturnum, from which significant suggestions can he wade concerning the study of photoperiodism as it relates to the initiation of floral primordia.

Materials and Methods

Cestrum elegans (Brongn.) Schlect. is a woody shrub which grows up to 3 m in height and produces clusters of red to magenta flowers (1). Native to Southern Mexico, it is grown as an ornamental shrub in various places, including Southern California. Material for this investigation was grown from the seed of plants established in the Los Angeles State and County Arboretum.

Cestrum nocturnum L. is a woody shrub which grows up to 3 m in height and produces clusters of greenish-white flowers having a strong night fragrance (1). Native of the West Indies and Central America including Southern Mexico, it is grown rather extensively as an ornamental shrub in Southern California. Material for this investigation, originally from plants in the Los Angeles area, are the plants used in previously reported work (2).

Plants used in this study, both the species and their hybrids, were grown in liter containers. Each individual plant was reused for various experiments by excising the shoot system and allowing a new shoot system to develop under the desired experimental conditions. Seedlings were grown to maturity on an 18-hour day, the shoot system then excised. and the plant placed tinder experimental conditions.

In addition, cuttings were made and used as test plants allowing for more than 1 set of data per clone in many cases. Under the conditions used, flowers opened in from 20 to 30 days after the floral primordia were initiated. The "days-to-bloom" is the time from excising the old shoot system and placing the plant under experimental conditions to the day the first flower opened. Flowers, in both the species and their hybrids, developed from buds in the axils of the leaves. In C. nocturnum they are distributed generally along the whole axis of the plant, whereas in C. elegans they are clustered near the tips of the branches, with a terminal inflorescence as the initial flower cluster.

Experimental Conditions. In all experiments temperatures were maintained at 23° from 0800 to 1600 and 19° from 1600 to 0800. Light conditions for the 8-hour photoperiod (short day, SD) were natural and extended from 0800 to 1600 for the 18-hour photoperiod (long day. LD) were natural supplemented in the morning and evening by florescent light at about 20 ft-c and extended from 0400 to 2200: and for the 24-hour photoperiod (continuous light. CL) were natural light from 0800 to 1600 and artificial light from florescent and incandescent sources at about 1000 ft-c from 1600 to 0800.

Hybridizing Techniques. Corollas and the attached stamens of flowers used as female parents were removed prior to the maturing of the anthers. As this time pollen from the male parent was placed on the stigma with a small brush. Flowers not hand pollinated on the female parent were continually removed until the fruit was harvested. No pollinating agents are present I the laboratories used, negating the necessity for bagging the flowers or using other such precautions, The hybrid plants are easily distinguished from the species by the color and shape of the corollas and fruit, and the shape and surface of the leaves.

Study Patterns. Five light pattern., were chosen for these studies as follows: SD, LD, CL, LSD (plants moved from an 18-hour photoperiod to an 8-hour photoperiod) and SLD (plants moved from an 8-hour photoperiod to an 18-hour photoperiod). These patterns were chosen because they give the maximum information about the photoperiod responses of C. nocturnum and give information concerning the day neutral (DN) condition. In both the LSD and SLD responses, 90-day-old shoot systems were transferred where possible. In hybrids whose LD and/or SD responses were such as to invalidate such a test 60 to 70-day-old shoot systems were substituted. In C. nocturnum shoot systems in this age range produced floral primordia when so treated (2).

Statistical Methods. In order to compare hybrid populations with the parent species as to the average length of days-to-bloom, the standard deviation and the unbiased estimate of the ninety-ninth percentile for the bloom period of the species was calculated by applying Student's t distribution (see table 1 for formulas). Hybrids that had not flowered by the end of the ninety-ninth percentile for the bloom period of the species are assumed to he significantly different. Wherever possible hybrids were grown for a minimum of 250 days and in some cases as long as 300 days if floral primordia had not been previously initiated. In all cases, these time limits were considerably longer than the ninety-ninth percentile for the bloom period of the species. Where bloom occurred during such a period beyond the ninety-ninth percent level it is so reported in the results.


Bloom Patterns of C. elegans. Mature plants were grown on SD, LD, and CL and their days-to-bloom established (table 1). There appear to be no significant differences in the length of days-to-bloom among the 3 conditions. Plants were tested for the possibility of this species responding to the LSD or SLD treatments by transferring them between the 3 photoperiods used at various ages. Floral primordia initiation did not occur as a result of these changes. indicating that this species does not respond to a LSD or SLD stimulus. Finally, plants were grown on 4-hour, 12-hour, 16-hour, and 20-hour photoperiods in addition to the above and in all cases the plants bloomed. These data establish C. elegans as a DN species.

Bloom Patterns of C. nocturnum. An extensive study of the bloom patterns and floral primordia initiation of this species has been reported (2). It was shown that this Species responds to a LSD stimulus with a critical day length) between 12 and 16 hours and a LD stimulus with a critical day length between 8 and 12 hours. In addition, it responds to a SLD stimulus at all age younger than the mean minus the ninety-ninth percentile of the days-to-bloom time period for the LD response. It has been deduced that this variation may he due to the accumulation of substances associated with floral primordia initiation on LD during a period when the plant is on a noninductive day length for that factor (2). Front these studies it appears that C. nocturnum is a LD and LSD plant. Assuming for purposes of this work that the SLD response is a result of the same floral primordia initiation pattern as the LD response, it will he used to test for the presence of the LD pattern in the hybrids studied thus preventing the inclusion of factors inherited from C. elegans which may cause floral primordia initiation on LD.

F1 Generation Hybrids. Although plants of this group exhibited a wide range of vigor in comparison to the parent plants, there appeared to he no relation between the vigor of the plants and their ability to initiate floral primordia. One difference that may have existed between the species population and the hybrid population studied is that plants were taken from a large seedling population of the species and in such a selection less vigorous plants could have been ignored inadvertently whereas all hybrids that could be brought to maturity were studied.

Table 1. Time-to-bloom in Days under Various Photoperiodic Patterns for the
Cestrum elegans, Cestrum nocturnum, and Their F1 Hybrids

Number of plants tested (n) in parenthesis.

Mean* 99%
Mean* 99%
Mean 99%
C. elegans (10) 122 ± 23 190 133 ± 36 240 103 ± 24 174
C. nocturnum (14)     137 ± 14 189    
C. elegans 9 X C. nocturnum (7) 124 ± 23 196        
C. nocturnum 9 x C. elegans (10) 162 ± 43 279        
* Mean ± standard deviation.
** Ninety-ninth percentile for population bloom period / where t99 is taken from tables for Student's t distribution.

Table II. Floral Primordia Initiation of Cestrum elegans, Cestrum nocturnum, and Their Various Hybrids

SD 20 20 20 0 10 8 7 7 11 9        
LD 20 20 20 20 13 1 11 5 19 15        
CL 20 20 20 20 8 7 8 3 18 12        
LSD 20 0 20 20 20 20 25 25 22 15 31 31 15 7
SLD 20 0 20 20 8 0 10 0 12 2 17 3    

I = C. elegans, II C. nocturnum, III = C. elegans ♂ x C. nocturnum ♀, IV = C. nocturnum ♂ x C. elegans ♀, V= (C. elegans ♂ x C. nocturnum ♀) selfed, VI = (C. elegans ♂ x C. nocturnum ♀) ♀ x C. nocturnum ♂, VII = (C. elegans ♂ x C. nocturnum) ♀ x C. elegans ♂.

T = number of test plants, B = number of plants blooming within the ninety-ninth percent bloom period of the parent species, SD = short day (8 hrs), LD = long day (18 hrs), CL = continual light, LSD = long-short day, SLD = short-long day.

C. nocturnum (female parent) was crossed with C. elegans (male parent). Single clones were used for the parents.. All plants responded to LSD treatment in the manner of C. nocturnum. Hybrids with 3-month-old shoot systems given SLD treatment did not respond. At this age all plants of C. nocturnum tested responded by initiating floral primordia, in a quantity and manner similar to LSD treatment (table II). All 10 plants tested for SD response produced floral primordial, 8 of them within ninety-ninth percentile for the bloom period of both species. The hybrids were tested in a similar manner on LD and 3 of 13 produced floral primordia with only 1 falling in the ninety-ninth percentile for the bloom period of the parent species. On CL, 7 of the 8 plants tested produced floral primordia within the ninety-ninth percentile for the bloom period of both species: the eighth plant was maintained under these conditions for 300 days and did not flower (table II).

C. elegans (female parent) was crossed with C. nocturnum (male parent). Single clones were used for the parents. All plants were tested in the same manner as above and responded to LSD but did not respond to SLD (table II). Seven tested on SD produced floral primordia within the ninety-ninth percentile for the bloom period of the parent species. Eleven were tested on LD and 7 bloomed, 5 within the ninety-ninth percentile for the bloom period of both species. Eight were tested in CL and 3 produced floral primordia within the ninety-ninth percentile for the bloom period of both species (tables I and II).

F2 Generation. Plants in this group are from 2 parents from the F1 generation in which C. nocturnum was used as a female parent. The plants used were average for responses of the F1 generation, producing floral primordia on LSD response. SD and CL but not producing floral primordia within the ninety-ninth percentile for the bloom period of the parent species on LD. In this population 15 of 22 plants tested produced floral primordia on LSD and 2 of 12 tested produced floral primordia on SLD response (table II).

Table III. Response of Selected F2 Plants to Various Light Patterns
Illustrating Possible Range of Responses in Any One Individual

Clone no. SD LD CL LSD SLD
259 + + + + 0
260 + + 0 +  
327     + 0 +
330   + + + 0
264 +   0 + +
336 0 0 + 0  
267 0 0 0 0 0

+ = floral primordia initiated within ninety-ninth percent bloom period of parent species,
0 = floral primordia not so formed. Abbreviations the same as listed for table II.             

Few plants have been tested for all 5 responses to date and the numbers are too small to he significant. However, those tested show various combinations of the 5 factors, including 1 plant that did not produce floral primordia in any of the 5 categories. Typical plants, illustrating this range of responses, are listed in table III.

F1 Backcrosses to Parent Species. Single plants of the F1 generation were backcrossed to the parent species. The LSD and SLD responses tested for in the cross with C. nocturnum and the LSD response were tested for in the cross with C. elegans. All 31 plants tested for LSD responded positively, while 3 of 17 tested for SLD responded positively in the cross with C. nocturnum. Seven of 15 tested for LSD responded positively in the cross with C. elegans (table IT).


The above data suggest the presence of 2 independent genes or gene groups controlling floral primordia initiation in C. nocturnum. In the case of the LSD response, all F1 hybrids. 15 of 22 in the F. hybrids, all of the plants from the backcross to C. nocturnum and 7 of 15 from the backcross to C. elegans produced floral primordia. These data establish the presence of a single dominant gene controlling this response. In the LD response, as checked for by the SLD response, none of the F1 hybrids, 2 of 12 in the F2 hybrids, and 3 of 17 in the backcross to C. nocturnum produced floral primordia. These data suggest that this response is controlled by 1 or more recessive genes.

In this and the previous report (2) it has been shown that a large number of floral primordia are initiated when C. nocturnum is held on a noninitiating photoperiod, SD in the ease of the LD gene or gene group and LD in the case of the LSD gene. and subsequently transferred to an initiating photoperiod for that factor, suggesting the build up of a substance associated with the metabolic pathway involved on a noninitiating photoperiod. This is further substantiated by the nature of the LD response. In previous experiments (2) it was shown that a gradual increase in number of floral primordia occurred when C. nocturnum was maintained on a 16 or 18-hour photoperiod, while, when transferred from a 16 or 24-hour photoperiod to a 12-hour photoperiod, both patterns manifest themselves independently. The LD response produced floral primordia after the bloom flush initiated by the LSD response had occurred, and in a manner similar to those plants maintained on a 12-hour photoperiod. In view of the fact that a substance involved is accumulated on the initiating photoperiod of the other gene or gene group, it would appear that a different substance is being utilized in each case.

An examination of the hybrids, both F1 and F2, gives some insight into the pattern present in the DN plant, C. elegans. There is an almost complete lack of flowering on LD in the F1 generation, while the parent plants bloom readily on that photoperiod. There is, however, a high percentage of F1 plants that bloom on an 8-hour day, while C. nocturnum does not do so, and a substantially greater number on CL than on LD. The previously established nature of the gene action from C. nocturnum suggests that all of this floral primordia initiation is the result of genes from C. elegans. The presence of a single complex pattern, quantitative in nature, with a high reaction under SD and CL and a low reaction under LD or a series of different pathways controlled by genes or groups of genes is suggested. The independent assortment of these patterns in the F2 generation table III) suggests the latter and is in keeping with the data reported for C. nocturnum.

Toward the clarification of these points, and others which will arise, experiments are now in progress to bring forth more detailed information concerning the inheritance of factors affecting floral primordia initiation in C. elegans, C. nocturnum and related species.


The author wishes to express his appreciation to Dr. Harry A. Borthwick, Chief Plant Physiologist, Agricultural Research Service, U.S.D.A., for his reading this manuscript and his valuable comments, and to Dr. John H. DeHardt, Department of Mathematics, California State College at Los Angeles, for his consulting on the statistical methods and formulas.

Literature Cited