Floral meristem identity genes
The lfy ap1 double mutant is unable to specify floral development of the lateral meristem, and develops only shoot-like structures (Weigel et al., 1992; Ratcliffe et al., 1998) . The overexpression of AP1 or LFY results in conversion of the shoot apical meristem and lateral meristems into flowers, and causes early flowering when the transgenic plants are grown either in long or short days (Mandel and Yanofsky, 1995; Weigel and Nilsson, 1995) . Additional factors in addition to LFY and AP1 must be involved in conferring the competence to flower, because 35S::LFY plants show a much weaker phenotype under short days, and 35S::LFY 35S::AP1 plants still produce a vegetative rosette (Blazquez, 1997; Blazquez et al., 1997) .
Other genes that are known to play a role in floral meristem identity include the genes FRUITFUL (FUL), CAULIFLOWER (CAL), and APETELA 2 (AP2) (Jofuku et al., 1994; Kempin et al., 1995; Okamuro et al., 1997; Gu et al., 1998) . Mutations in CAL, which encodes a MADS box protein related to AP1, cause no phenotype but in a cal ap1 double mutant determinate floral meristems fail to develop, resulting in continued proliferation and a phenotype that has inflorescences similar to those of cauliflowers (Brassica oleracae) (Bowman et al., 1993; Kempin et al., 1995) . In addition, microarray analysis recently identified many genes whose expression is increased or decreased at the shoot apical meristem on the induction of flowering (Schmid et al., 2003) . Surprisingly, the majority of these were downregulated, and two AP2 domain-encoding genes, SCHLAFMUTZE and SCHNARCHZAPFEN, were shown to be active repressors of flowering. These data should provide access to a wide set of genes that act at the shoot apex during the floral transition.
Repression of the floral transition
The TERMINAL FLOWER 1 (TFL1) gene acts at the SAM to repress the floral transition, and maintain vegetative growth (Alvarez et al., 1992; Bradley et al., 1996; Bradley et al., 1997) . Mutation of tfl1 causes conversion of the SAM and axillary meristems to flower, similar to the phenotype of 35S::LFY plants (Shannon and Meeks-Wagner, 1991; Alvarez et al., 1992) . Furthermore, LFY is expressed ectopically in the SAM of tfl1 mutants. The role of TFL1 therefore seems to be as a repressor of LFY expression and of the floral transition. This is further supported by 35S::TFL1 plants which are very late flowering and similar in phenotype to lfy mutants (Ratcliffe et al., 1998) . TFL1 and the flowering-time protein FT are closely related in sequence suggesting that they might have antagonistic functions (Kobayashi et al., 1999) .
In the vegetative growth stage both TFL and LFY mRNA levels are low at the SAM, with transcription levels increasing during the transition to flowering. However, they are expressed in different regions of the meristem, LFY at the primordia and TFL1 in the centre, indicating that spatial location is important in the induction of meristem identity genes (Bradley et al., 1997; Ratcliffe et al., 1999).
Forward to Integration of flowering-time pathways