Cucurbit Genetics Cooperative Report 4:2-4 (article 1) 1981
M.D. Edwards and R. L. Lower
University of Wisconsin, Madison, WI 53706
Seeds from compact (cp cp) cucumber plants often germinate poorly when compared to those from normal vining-type (Cp Cp) plants, particularly in adverse environments.”Compact” seeds also appear somewhat unlike “vining” cucumber seeds. The average seed weight is about 35% that of normal seeds. Also, seeds are often more cylindrical than the flat, ellipsoidal shape characteristic of cucumber seeds.
This study was conducted to gain insight into the variability present for a number of seed characteristics in a heterogeneous population of compact genotypes and to determine if visually apparent seed characteristics were associated with poor germinability. Initially, a random sample of 62 open-pollinated, mature fruit was selected from a population of compact genotypes at Hancock Experimental Farm in September 1979. Fruit length and diameter were measured, then seed were extracted and fermented at room temperature for three days to facilitate cleaning. Washed seed were air-dried for thee days, then evaluated for total seed number and percentage of “flat” seeds prior to packaging. Seeds of three fruit were discarded at this time due to failure of seed development.
Three months later two samples each of 10 “flat” and 10 “cylindrical” seeds were taken from each of the remaining 59 seed lots and evaluated for sample weight. There samples were then planted 1.5 cm deep in vermiculite in wooden flats on a greenhouse bench and overhead irrigated until germination occurred. Experimental design was a randomized complete block with two replications. Emergence percentage and rate (mean number of days to emergence) were recorded for each sample. In addition, visual ratings were recorded for severity of cotyledonary abnormalities and vigor at the time the second true leaf was beginning to expand.
A wide range of variability was observed for fruit and seed characteristics among 59 randomly-sampled fruit as indicated in Table 1. Partitioning of the observed variability for emergence percentage is shown in Table 2. Most of the variability was observed between fruit samples, but highly significant components were also attributable to replicates and “seed types” within each fruit. Phenotypic correlations between seed and fruit characteristics are presented in Table 3. Emergence percentage is significantly and positively correlated with percentage flat seeds, total seed number, and seed weight.
Several multiple regression models were fitted in an attempt to predict emergence percentage from seed characteristics. Although a number of seed characteristics are correlated with emergence percentage, these characteristics are significantly intercorrelated and as a result account for some of the same variability in emergence percentage. The best model accounted for only 24% of the variability observed for emergence percentage in this sample of fruit. Presumably, factors other than visually apparent seed characteristics are important in determining seed quality in compact cucumbers.
Table 1. Distribution of fruit and seed characteristics for 59 randomly-sampled fruit.
Characteristic |
Mean |
S. D. |
Minimum |
Maximum |
| Fruit length (cm) | 14.5 | 2.9 | 9.0 | 22.0 |
| Fruit width (cm) | 6.7 | 0.62 | 5.5 | 8.0 |
| Seed number | 162.0 | 65.0 | 28.0 | 313.0 |
| Percent flat seeds | 58.0 | 22.0 | 0.0 | 97.0 |
| Seed weight (g/seed) | 0.011 | 0.002 | 0.007 | 0.016 |
| – Flat seeds | 0.011 | |||
| – Cylindrical seeds | 0.011 | |||
| Emergence percentage | 42.0 | 30.0 | 0.0 | 98.0 |
| – Flat seeds | 48.0 | |||
| – Cylindrical seeds | 35.0 | |||
| Days to emergence | 7.6 | 1.6 | 5.8 | 12.3 |
| – Flat seeds | 7.6 | |||
| – Cylindrical seeds | 7.6 |
Table 2. Analysis of variance for emergence percentage among 59 random fruit.
Source |
df |
Mean squares for emergence |
| Fruit | 25 | 0.354** |
| Replication | 1 | 0.183** |
| Error 1 | 59 | 0.015 |
| Seed type within fruit | 59 | 0.043** |
| Error 2 | 59 | 0.009 |
| R² = 0.94 |
** significant at the 0.01 level.
Table 3. Phenotypic correlations between fruit and seed characteristics among 59 random fruit.
Trait |
Fruit diameter |
% flat seeds |
Seed number |
Seed weight |
Emergence % |
Days to emergence |
| Fruit length | 0.21 | 0.16 | 0.50** | 0.32** | 0.01 | -0.19 |
| Fruit diameter | -0.04 | 0.06 | 0.49** | -0.16 | 0.14 | |
| % flat seeds | 0.40** | 0.27** | 0.45** | -0.45** | ||
| Total seed number | 0.06 | 0.34** | -0.47** | |||
| Mean seed weight | 0.31* | -0.13 | ||||
| Emergence % | -0.60** |
*,** significant at the 0.05 and 0.01 levels, respectively.