Southern blot analysis indicates that the rice genome contains single copies of genes encoding type A (phyA) and type B (phyB) phytochromes. We have isolated overlapping cDNA and genomic clones encoding the entire phyB polypeptide. This monocot sequence is more closely related to phyB from the dicot, Arabidopsis (73% amino acid sequence identity), than it is to the phyA gene in the rice genome (50% identity). These data support the proposal that phyA and phyB subfamilies diverged early in plant evolution and that subsequent divergence accompanied the evolution of monocots and dicots. Moreover, since rice and Arabidopsis phyB polypeptides are more closely related to one another (73% identity) than are monocot and dicot phyA sequences (63-65% identity), it appears that phyB has evolved more slowly than phyA. Sequence conservation between phyA and phyB is greatest in a central core region surrounding the chromophore attachment site, and least toward the amino-terminal and carboxy-terminal ends of the polypeptides, although hydropathy analysis suggests that the overall structure of the two phytochromes has been conserved. Gene-specific Northern blot analysis indicates that, whereas phyA is negatively regulated by phytochrome in rice seedling shoots in the manner typical of monocots, phyB is constitutively expressed irrespective of light treatment. In consequence, phyA and phyB transcripts are equally abundant in fully green tissue. Since Arabidopsis phyB mRNA levels are also unaffected by light, the present results suggest that this mode of regulation is evolutionarily conserved among phyB genes, perhaps reflecting differences in the functional roles of the different phytochrome subfamilies.