Quantitative cytogenetics of polyploids

Abstract

In the 62 years that have passed since H. Kihara and T. Ono defined autopolyploidy as doubling of the same chromosome set, there is still a lack of understanding of the term. The causes of confusion are several: a lack of cytogenetic knowledge by many who use the term, a lack of quantification of cytogenetic data, the attempt by many systematists to use taxonomic categories for what is basically a genetic phenomenon, and until 1982 the lack of suitable mathematical models and equat; Jns that can provide expected frequencies of meiotic ~onfip.~rations in autopolyploids with known chiasma frequencies. It is necessary to point out that H. Kihara and T. Ono studied only diakinesis or metaphase stages in their classical analysis of chromosome pairing, and today one still reads articles referring to chromosome pairing at diakinesis or metaphase I stages. I have pointed out elsewhere that examining only metaphase I stages can give a misleading impression of the actual pairing configurations at pachytene. If indeed a polyploid has identical sets of chromosomes, it should be possible to mathematically model such a system and to derive equations that predict with great accuracy the kinds and numbers of expected meiotic configurations at pachytene and metaphase I. Organisms that deviate significantly from expected configurations for autopolyploids are then allopolyploids. Not all allopolyploids will be strict bivalent formers. In nature, there should be a series of allopolyploid types extending from those that have slightly differentiated genomes and various frequencies of multivalents to those that have only bivalents. Those allopolyploids with multivalents can have different adaptive strategies depending on the environmental conditions that existed at their inception or later because recombination between the different genomes can produce new genotypic combinations not found in strict allopolyploids. Likewise, we should be able to synthesize various types of polyploids for specific agronomic purposes and for different environments. This workshop emphasized quantitative cytogenetic methods in the analysis of both synthetic and natural autopolyploids. The four principal speakers in sequence were R. C. Jackson, D. P. Hauber, B. G. Murray, and J. Kuspira. The development of autopolyploid pairing and chiasmata distribution models and equations derived from them were described by R. C. Jackson who emphasized that essential data for a quantitative -analysis of autopolyploids were the following: basic chromosome number, maximum and mean chiasmata numbers per bivalent or bivalent equivalent, mean number of chiasmata, per meiocyte, and any significant variation in chiasma frequency among bivalents. This latter point should be emphasized because some workers who have used the models and equations have assumed a priori that all bivalents behave in the same way. That this is not true was pointed out by D. P. Hauber in his analysis of Haplopappus. Bivalents with different chiasmata forming capabilities can be treated as subclasses and then the total configurations can be summed. With the exception of many studies of agronomic and a few wild species, most studies of polyploids have not met the minimum cytogenetic standards for a rigorous quantitative analysis.