36 



Centre in Norwich, England, and Elliot M. 

 Meyerowitz of Caltech. The evidence for the model 

 came from genes identified in two distantly related 

 species, arabidopsis and snapdragon. (Arabidopsis is 

 the plant biologist's counterpart to the zoologist's 

 fruit fly, the most important plant studied in the lab- 

 oratory as a model. In 2000 it became the first plant 

 to have its genome sequenced.) 



According to the ABC model, genes that, col- 

 lectively, carry out three functions, A, B, and C, 



Oddly, the species that serves as an experimental 

 model for the entire plant kingdom appears to be 

 unique in the way it makes its flowers. 



must all be active in the floral meristem to form 

 the four kinds of flower organs. Genes that carry 

 out function A act in the sepal and petal whorls of 

 the meristem; genes that carry out function C act 

 in the stamen and carpel whorls; and genes that 

 carry out function B straddle the two, acting in the 

 petal and stamen whorls. Thus, A genes form 

 sepals, A and B genes together form petals, B and 

 C genes together form stamens, and C genes form 

 carpels [see diagram on page 39]. 



That's how the ABC model answers the evolu- 

 tionary botanist's first question, What are the basic 

 instructions for flower-organ formation? It also ad- 

 dresses parts of the second question: What are the 

 variations in the instructions for diverse kinds of 

 flowers? Although it does not explain variation in 

 flower size, color, or symmetry, it does explain vari- 

 ation in flower form. 



For example, the ABC model predicts that if a 

 plant's C genes are inactivated, its A genes will be- 

 come active throughout the developing meristem. 

 With no C genes, stamens and carpels cannot 

 form, but additional whorls of petals and sepals 

 will take their place as the A and B genes exert 

 their effects on the inner two meristem whorls. C 

 genes also tell the flower to cease making new or- 

 gans; without C genes, the pattern of sepal-petal- 



Arabidopsis thaliana (rightmost flower at right) is the plant of choice for labora- 

 tory study, the "fruit fly" of the plant kingdom. Its small size, short, six-week life 

 cycle, and simple genome make it relatively easy for investigators to manipulate. 

 All plant species possess genes related to an arabidopsis gene called AP1. Those 

 genes may have played an important role in the evolution of flowers with organs 

 arranged in concentric whorls, such as arabidopsis and morning glory (far right of 

 opposite page), from flowers with spirally arranged organs, such as water lily 

 (center flower at right). Snapdragon (leftmost flower at right) is another com- 

 monly studied plant. Genetic evidence from snapdragon and arabidopsis under- 

 lay the first widely accepted genetic model of how flower organs develop, the 

 ABC model. The flowers are not all shown to the same scale. 



NATURAL history June 2006 



petal may repeat itself several times. That is exactly 

 what takes place when the C genes in arabidopsis 

 are experimentally inactivated. Cultivated roses 

 and carnations, too, have extra whorls of petals in 

 place of at least some of their reproductive organs, 

 and plant biologists hypothesize that the change 

 results from inactive C genes, in line with the 

 ABC model. 



Changes in the activity of A, B, and C genes 

 have also been posited to explain other flower 

 forms that don't conform to the standard 

 sepal-petal-stamen-carpel construction. 

 Lilies and tulips, for instance, have two 

 whorls of petals instead of one whorl of 

 sepals and one whorl of petals. The ABC 

 model explains the double petals as a mu- 

 tation in which B genes become active in 

 the outer whorl of the meristem, where A genes 

 usually act alone. When A and B genes both act in 

 the outer two regions of the meristem, two whorls 

 of petals form. 



Explaining flower formation in two distantly re- 

 lated species with strikingly different flowers — 

 snapdragons are showy, tubular, and 

 bilaterally symmetrical, whereas 

 arabidopsis flowers are 

 inconspicuous, dish-shaped, 

 and radially symmetrical — 

 was an impressive 

 achievement. That a single 

 model could account for 

 flower development 

 in two plants so differ- 

 ent from each other gave it 

 great weight. And the ABC 

 model has proved an in- 



X 



