Probably because of their fundamental value in agriculture, many angiosperms have been chosen for genome size studies since the 1950s. Consequently, angio-
sperms are probably the most studied higher order group with respect to genome size. Nevertheless, due to the large number of species recognized (at least 250 000) this corresponds to only ca. 1.8%. Whilst early studies tended to concentrate on temperate and crop plants, since 1995 there has been increased awareness of the need to make genome size data more representative of angiosperms as a whole. Improvements have been made in the systematic coverage following the active targeting of key systematic gaps identified at the first and second Plant Genome Size Meetings held at the Royal Botanic Gardens, Kew in 1997 and 2003, and by the formal constitution of an international group for genome size analysis named GESI (GEnome Size Initiative) to drive forward this vital process over the coming years (Bennett and Leitch 2005d). For example, at the 2003 meeting, with species representation standing at ca. 1.6% (i.e. ca. 4200 species), a specific target of estimating a further 1% (i.e. 2500) species in the next 5 years was set. In the most recent analysis, output of C-values was shown to be at a record high (ca. 290 first estimates for species per year; Bennett and Leitch 2005b). Species representation is thus expected to increase significantly over the next 5 years. At the family level, representation has also improved, increasing from 30 to over 50% between 1997 and 2003, largely due to targets set at the first Plant Genome Size Meeting (Hanson et al. 2001a, 2001b, 2003, 2005). At the genus level, over 8% (1126) of the 14 000 genera now have at least one C-value and this is targeted to rise to 10% by 2009 and may reach 15% by 2015. Moreover, generic representation may approach 100% in monocots as they are targeted for holistic genomic studies (including C-values) for the global Monocot Checklist Project (Govaerts 2004).
Representation of genome size data in some other key categories, however, remains poor. For example, despite highlighting the need to improve geographical representation over a decade ago (Bennett and Leitch 1995), gaping chasms still remain. With some exceptions, the sample is still dominated by crops and their wild relatives, model species grown for experimental use and other species growing near laboratories in temperate regions mainly in Western Europe and North America. There is still a dearth of data from taxa endemic to China, Japan, South America, and Africa. Similarly, although island floras are known to be rich in endemics, there has been no publication reporting C-values for any large island such as Borneo, New Guinea or Madagascar where 80% of the 12 000 described plant species are endemic (Robinson 2004). Other plant groups that were identified as poorly represented in terms of genome size include taxa from bog, fen, tundra, alpine, and desert environments and specific life forms such as parasitic, saprophytic and epiphytic species and their associated taxa.
Was this article helpful?