How Much Did Crop Diversity Decline in the 20th Century?

It is generally assumed that crop diversity declined dramatically during the 20th century. This trend is blamed upon market pressures and the rise of corporate agriculture, among other things. But is the underlying assumption accurate? Paul Heald and Susannah Chapman of the University of Georgia (law and anthropology, respectively) suggest we may need to rethink what we think we know about vegetable crop diversity. In a new paper, "Crop Diversity Report Card for the Twentieth Century: Diversity Bust or Diversity Boom?", they present evidence that crop diversity has not declined meaningfully at all.

According to the conventional wisdom, the twentieth century was a disaster of monumental proportions for vegetable crop diversity. The conventional wisdom is wrong. Our study of 2004 commercial seed catalogs shows twice as many 1903 crop varieties surviving as previously reported in the iconic 1983 study on vegetable crop diversity. More important, we find that growers in 2004 had as many varieties to choose from (approximately 7100 varieties among 48 crops) as did their predecessors in 1903 (approximately 7262 varieties among the same 48 crops). In addition, we cast doubt on the number of distinct varieties actually available in 1903 by examining historical sources that expose the systematic practice of multiple naming. Finally, by looking more closely at the six biggest diversity winners of the twentieth century (tomatoes, peppers, lettuce, garden beans, squash, and garlic), we suggest that patent law is virtually irrelevant.
These results will be a surprise to many, but I think they are largely consistent with what economic theory would predict. While there are pressures toward greater commodification and standardization of agricultural goods, there are also pressures to satisfy the demands of niche producers and consumers. As our ability to modify crop varieties improved, and the costs of customization declined, desired crop varieties should proliferate.

These trends might not occur in tandem. That is, the pressure toward commodification to increase productivity and reduce costs may precede the pressure toward the development of specialty crops. If so, then one would expect to see an initial decline in crop diversity before an eventual increase. (Think of something like the agricultural equivalent of the "Environmental Kuznets Curve.") Some crop varieties are lost along the way, to be sure, but new varieties emerge as well, and I see no reason to presume that the older varieties are necessarily superior than the new ones. In many cases they will each have been the product of human design.

Consider this example. During the initial "Green Revolution" there was a decline in crop diversity as it was too costly to breed desired traits into all existing varieties of individual crops. Increased agricultural productivity came at the expense of local crop diversity. Over time, however, more advanced techniques make it much less expensive and time consuming to insert a desired characteristic into a given plant. This has enabled the insertion of desired traits into traditional local varieties, so that crop quality can be improved without sacrificing crop diversity. Somewhat ironically, some of the same groups that complain the most about the loss in crop diversity also oppose the technologies that can enable agriculture to meet more human needs without sacrificing it.

UPDATE: Study co-author Paul Heald adds some thoughts in the comments.