American Rose Annual 44: 141-145 (1959)
Roses Have Satellites, Too
Edward B. Risley*
Durham, N.H.

*Assistant horticulturist, University of New Hampshire.
Published with the approval of the director of the New Hampshire Agricultural Experiment Station as scientific contribution number 224.

WHILE the geophysical scientists are launching earth satellites, recent examination of the chromosomes in the cells in rose plants has revealed the presence of satellites there also. This discovery resulted from an attempt to count the chromosomes in the root tips of some of the roses that have resulted from seven years of work at the University of New Hampshire in breeding hardy Ramblers suitable for northern New England.1

Chromosomes have a rod-shape only for a short period of time (called the metaphase) when each cell is newly-formed and while the slender threads of genes (the tiny bodies believed to control hereditary characteristics) are tightly wound into a hollow spiral coil (somewhat resembling a screen-door spring) which stains black and looks like a rod.

Each rose variety has a definite number of chromosomes in all its cells. Seven chromosomes make a set in Rose, and most roses that are fertile have two sets (diploids) or four sets (tetraploids). Those having three sets (triploids) are mostly sterile. Table 1 lists examples of these normal conditions, and also some unusual ones such as fertile triploids, sterle diplolid and sterile tetraploids.

Rose chromosomes are very small. Most of the ones illustrated here are two to three microns long, which means that it would take around 1300 of them laid end to end to equal 1". We hoped to see the shape of each individual chromosome clearly enough to be able to identify it as an individual and trace its appearance in the cells of its "children," "grandchildren," etc., by magnifying it 1250 times under an excellent research microscope. In Table 1, an asterisk marks those rose varieties whose root-tip cells have one chromosome bearing a satellite (a short coil of chromatin material attached to the chromosome by a chromatin thread) . These "markers" aid greatly in identifying individual chromosomes, and since the chromosome is considered one of the most fundamentally-basic units in a plant, taxonomic studies of the true relationships of the varieties of a species, or species of a genus, are greatly enhanced by chromosome identification. In Rosa this would appear to be a most difficult, but not impossible task.

Figure 1
Figure 2
Skinner's Rambler. The 14 chromosomes in somatic metaphase appear to be separated into 7 pairs.   Belinda. One of the 14 chromosomes has a satellite.

Figure 3
Figure 4
N.H.# 5340  (Skinner’s Rambler x Belinda). The satellite was retained in the hybrid   N.H. #535 (Skinner’s Rambler x Belinda). No satellite appears in this hybrid.

Figure 5
Figure 6
Max Graf (Rosa wichuraiana x Rosa rugosa). One of the 14 chromosomes in this nearly-sterile diploid has a satellite. Note groups of three chromosomes at center and bottom.   N.H. #545 (Skinner’s Rambler x Max Graf). The Max Graf satellite and two groups of three chromosomes were retained in this hybrid. Note the ring-shaped chromosome held in that position by stands of chromatin material.

Figure 7
Figure 8
Gruss an Aachen. A fertile triploid with one satellited chromosome and a ring or two held in position by a chromatin strand.   N.H. SR3GAA (Skinner’s Rambler x Gruss an Aachen). First generation shows 14 chromosomes and a very small satellite.

Figure 9
Figure 10
N.H. #551 (Skinner’s Rambler x Gruss an Aachen). Second generation is a fertile diploid with satellited chromosome.
  Yellow Pinnochio. A fertile tetraploid with 28 chromosomes.

Figure 11
Figure 12
N.H. #5425 (Skinner’s Rambler x Yellow Pinocchio). A sterile triploid with 21 chomosomes. Only 33% of the cells examined had all 21: 4 had 20, 7 had 19, 6 had 18 and 3 had only 17.   N.H. #500 (Rosa wichuraiana x Betty Bland). The nature of sterility in this diploid may be the same as that in Max Graf.

Figure 13
Rosa carolina plena (Lynes) A sterile tetraploid with 28 chromosomes.    

Table 1. Some Chromosome Counts in Garden Roses and Their Relation to Breeding Behavior
1. Fertile Diploid x Fertile Diploid
Skinner's Rambler x Belinda*

Fertile Diploid
N.H. #5340* and N.H. #535
2. Fertile Diploid x Highly Sterile Diploid
Skinner's Rambler x Max Graf*

Fertile Diploid
N.H. #545*
3. Fertile Diploid x Fertile Triploid
Skinner's Rambler x Gruss an Aachen*

N.H. SR3GAA* .. F1
N.H. #551* F2
4. Fertile Diploid x Fertile Tetraploid
Skinner's Rambler x Yellow Pinocchio
Skinner's Rambler x Rosa virginiana

Sterile Triploid
N.H. #5425
N.N. #5314
5. Sterile diploids
Mrs. Anthony Waterer? (*variable)

(Rosa rugosa x General Jacqueminot)
(Rosa wichuraiana x Betty Bland)
6. Sterile tetraploid
Rosa carolina plena (Lynes)2

*One chromosome has a satellite in these roses.

Figures 5 and 6 show the satellited chromosomes and two groups of three others connected by "threads" in Max Graf and its seedling, N.H. #545. Such a relationship is not constant in all cells examined in these roses, and a great amount of observation would be essential before positive identification of any one chromosome could be made in Rosa.

Of particular interest is the presence of chromatin threads connecting Rosa chromosomes at metaphase in root-tip cells. In Figure 6, a chromosome is bent into a ring and held fast to another chromosome by these threads. In Figure 7, a ring of two chromosomes is held by a thread. In many of the other figures, groups of three or more chromosomes appear to be interconnected.

This condition is normally only associated with cell division in the formation of pollen and embyro sacs (in which these connectives are called chiasmata and the resultant chromosome groups are termed bivalents, trivalents, quadrivalents, etc.). Such groupings are associated with malfunction and sterility in varying degrees. What effect these threads have in root-tip cells is unknown.

The number of chromosomes in the cells of one plant are supposed to be constant, and almost always are so. The full compliment of 21 chromosomes of the triploid offspring of Skinner's Rambler x Yellow Pinocchio are shown in Figure 11. All 21 were not always present in the root-tip cells. Only 33% contained all 21, while the remainder had 20, 19, 18 and 17 chromosomes per cell. This rose should produce an occasional mutation or sport from a bud if this condition of variable chromosomes number exists in all the tissues of the plant. [CybeRose note: Or if plants were regenerated from root cuttings.]

  1. Risley, Edward B. Breeding Winter-Hardy Rambler Roses, American Rose Annual 42:57-73 (l957).
  2. Lynes, Doris and Wilson. A Rediscovery of the Double Form of Rosa Carolina. American Rose Annual 40: 19-24 (1955).

Risley: Male Controls Sprouting (1958)