The American Rose, 1965 p. 8, 9, 30
About Rose Seed Germination
R. N. Stewart and Peter Semeniuk, Beltsville, Maryland.
Horticulturists, Crops Research Division, Agricultural Research Service,
U. S. Department of Agriculture, Beltsville, Maryland.
(This work, and the rose breeding program, at Beltsville, Maryland, have been supported for a number
of years by annual grants from the American Rose Foundation which is gratefully acknowledged).

MOST ornamental plant growers are interested in new and better rose varieties. Poor seed germination is one of the major problems limiting the breeding of the rose. A description of some recent experimental work on the germination behavior of rose seeds will serve to illuminate the kind of long and careful background research which may be required in a rose-breeding program.

The successful pollination and the development of a well-filled fruit is only the beginning in breeding for a new rose variety (Fig. 1), The experienced hybridizer knows that these precious seeds are dormant and that most of them will never germinate without special care. With some rose varieties frequent sequel to pollination is a structure which is unsaleable and unfruitful (Fig. 2).

Germination of seed formed by some of the best rose varieties is so low that they are excluded from breeding programs. Low germination keeps the breeder from observing and measuring the full range of the variability that would appear in his seedlings.

Fundamental mechanisms of seed dormancy are far from being understood. The regulation of the various physiological and biochemical changes which occur prior to and during germination presents a highly complex set of problems. These have been the subject of an enormous amount of research for many years; roses have been included in these studies. The usual approach has been to manipulate the environment of the rose seed during the development on the mother plant and after harvest. The usual factors considered have been moisture availability, composition of the air, light, and temperature.

While the rose "seed" is a fruit, called an achene, it does exhibit the properties and processes associated with true seeds and will be considered a physiologic seed. The thick, hard, outer coat is derived from the ovary wall. The thin, leathery, inner coat is from the ovule walls or integuments. The rest of the seed is the mature, fully-developed embryo with two large cotyledons and a relatively small axis of stem and root initials.

Rose seeds harvested from mature fruits are dormant. In our experience at Beltsville the seeds from many different species and garden roses have failed to germinate when planted in a moist medium and held at a temperature of 65 degrees F or above. Several exceptions to this general occurrence of dormancy appear in the literature.

Crocker and Barton at the Boyce Thompson Institute reported that after-ripening rose seeds at 40'F and planting them in a moist medium resulted in greatly increased germination. We have confirmed this observation. Low temperature after-ripening occurs only if the seed is moist; and germination trials with seed from hips harvested at intervals through the entire winter from outdoor plantings show that after-ripening does not occur in the hip on the plant. In fact, seed collected from over-wintered hips and after-ripened at 40 degrees do not germinate as well as seed collected in the fall.

There have been frequent reports that passage of seed through various animals or birds improves their germination. These reports prove that rose seeds are tough and pass through most fruit eaters unharmed, but still dormant. Following a late spring snow during a robin migration in Beltsville last year, we were able to observe intensive feeding for several days on over-wintered fruits from a hedge of Rosa multiflora. Germination tests of seeds from robin droppings and of seeds from fruits on the hedge indicated that the after-ripening requirement was the same.

In many kinds of plants whose seed is dormant when mature, immature seed from green fruit will germinate immediately. This is not true with rose seed. In one small way, rose seed has proved amenable. When harvested and removed from mature fruits, it can be stored dry at room temperature for long periods with no apparent effect upon its physiologic condition. We have found no change in per cent of germination or after-ripening requirement from prolonged storage. This characteristic has enabled us to conduct experiments, through a whole season, with samples of seed from a single large lot. This assures relatively uniform material for the several tests and makes it possible to compare results of the several tests with more confidence.

At Beltsville we find that open-pollinated seed produced by the same plant in successive years can differ greatly in the per cent of germination after a standard after-ripening period. Von Abrams in California made similar observations, and felt that this was due to a difference in temperature from one year to the next during the maturing of the fruit. However, in experiments where fruit was ripened on the plants at a series of controlled temperatures we found no difference in after-ripening requirement or in germination.

In our fields we feel that at least part of the variability in germination from year to year is due to seasonal fluctuations in numbers of several kinds of insect pests which parasitize rose seeds, and which are widely distributed in temperature zones. We have found the rose chalcid larvae in as much as 50 per cent of the seed of several species the last two years, with apparent differences in frequency between seed from different locations and different years. Eggs are deposited in the very young seeds and larvae feed on the embryo, eventually consuming it and occupying the same space in the seed. Seed with larvae and seed with embryos cannot be differentiated by microscopic examination or by flotation. They were discovered when excising embryos for nutrient culture experiments. Seeds with aborted embryos are usually separated by flotation, but a true estimate of the germination potential of a lot of seed can be obtained only by carefully cutting open a representative sample. Without this information the effectiveness of the treatments is difficult to assess.

The embryos of mature rose seeds will grow if excised and cultured on nutrient-medium. The primary dormancy of rose seeds is therefore assumed to be imposed by the seed coat. Attempts to break the primary dormancy by acid treatment, or by scarification have failed. Complete or nearly complete removal of both the outer and inner seed coat is necessary to break primary dormancy.

The compensating temperature can be defined as that temperature at which moist seed will neither after-ripen nor move toward secondary dormancy. Rose species with a short after-ripening requirement have been found to have a high (above 75°F) compensating temperature and those with a long after-ripening requirement a low (below 65°F) compensating temperature.

The after-ripening requirement of the various rose species tested has ranged from thirty days to three hundred days. If the after-ripening period is interrupted by intervals at above the compensating temperature, a secondary dormancy is often established in the embryo itself. When such embryos are excised they will not grow. Breaking this secondary dormancy apparently requires after-ripening at 40°F for even longer intervals than the primary dormancy. The longer the after-ripening requirement of seeds of a rose species, the more sensitive it is to high temperature interruptions, i.e., shorter intervals at lower temperatures will impose secondary dormancy of the embryo. Conversely, seed of species with a short after-ripening requirement are put into secondary dormancy only by longer intervals at higher temperatures.

Several experiments have measured the effect of the temperature following after-ripening upon germination. Up to a certain point, the only effect of increase in temperature is to speed up the germination with no effect upon the total per cent of the germination achieved. Seeds held at 40 degrees F eventually germinate to the same degree as seeds adequately after-ripened and moved to 65 degrees F or 75 degrees F. Temperature above 75 degrees F can reduce germination, especially of seeds with a minimum of after-ripening and of a species with a low compensating temperature.

Considerable data has been collected on the germination response of the seeds of a number of rose species and garden roses to exposure to various temperature regimes. Within species, the responses are generally quite consistent, the species can be separated into several groups with quite different responses. The response of seeds of hybrid origin is quite variable. We hope to correlate temperature responses during seed germination with the temperature responses required for the subsequent initiation and development of flowers. Such information about ancestral species seems likely to shed some light on the behavior of the modern garden and florist rose.


  1. Harvest the fruits when they turn color; separate the seed and plant in sphagnum moss.
  2. Keep moist and store at 40 degrees F until about 5 per cent germination occurs at that temperature, then bring into a 65- to 75-degree temperature. Germination will be complete in fifteen to twenty days. The amount of germination will be close to the maximum potential germination.
  3. Further after-ripening of the ungerminated seed is not worthwhile because a secondary dormancy will probably have been imposed.
  4. The length of the original after-ripening is safely beyond the minimum requirement by the time 5 per cent germination occurs at 40 degrees.
  5. The old practice of putting the old seed flats back in cold storage a number of times was not very successful. We would recommend starting a new lot of seed.