Many of the stories that have appeared in this column recount the way in which the timely appearance of a paper catalyzed the pursuit of a new line of research. The experience is, happily, not limited to new papers. One of the greatest pleasures in research is to discover (or more accurately, to rediscover) a gem of a paper that has been in the literature for some time but whose findings or promise have not yet been realized.The rediscovery of Mendel's work or the re-examination of Walcott's interpretations of the Cambrian fauna of the Burgess Shale are famous examples of this, but our more modest excursion from the comfortable confines of Drosophila melanogaster developmental genetics to the psychedelic world of butterfly wing patterns was also triggered by such a rediscovery.In the early 1980s, I was prompted to work on the developmental genetics of Drosophila melanogaster by the conviction that the way to achieve an understanding of morphological evolution was through the genetic basis of animal form. Yet, for the next decade, I did not initiate any comparative or evolution-oriented work. In part, my hesitation was due to a sense that, because we had so much to learn about developmental regulatory mechanisms in model systems, comparative studies would be superficial at best. Part of the problem, however, was my failure to identify tractable models for such comparative work.A casual conversation on a seminar visit at Duke University in 1991 turned things around. While hurrying across a parking lot to make my next appointment, Fred Nijhout asked me, quite innocently, whether the mechanisms we were studying that position bristles on the fruitfly could help explain what really interested him — the formation and diversity of butterfly wing color patterns.“I don't know,” was my lame answer but I promised Fred I would think about it. As I knew nothing about butterflies (I was not a collector as a kid because houseflies were about the most exciting fauna in Toledo, Ohio), I went to the literature to find out about their wing patterns. Much of what had been published, I learned, had been written by Fred.I found a pair of papers written in 1980 that describe some remarkable manipulations of butterfly wing patterns [[1]xPattern formation on lepidopteran wings: Determination of an eyespot. Nijhout, HF. Dev Biol. 1980; 80: 267–274CrossRef | PubMed | Scopus (103)See all References, [2]xOntogeny of the color pattern on the wings of Precis coenia (Lepidoptera: Nymphalidae). Nijhout, HF. Dev Biol. 1980; 80: 275–288CrossRef | PubMed | Scopus (46)See all References]. The central interest was in eyespots, the variously colored concentric rings of pigmented scales that are used by Lepidoptera to avoid predation. Fred surmised that these rings were organized by a focal source at their center so he tested this idea by ablating small areas of the developing wing. Sure enough, the presumptive centers were necessary for the formation of the eyespots. More spectacularly, when Fred transplanted these centers to other sites on the wing, new eyespots were induced.Figure 1Wing walking: a fruit fly strolls across one of the butterfly's best works of development.View Large Image | View Hi-Res Image | Download PowerPoint SlideThe eyespot center, or focus, was clearly a developmental organizer on a par with more famous regions or zones of amphibian embryos or vertebrate limb buds. Moreover, the organizer and eyespot patterns had evolved within this order of insects. So, Fred's experiments had identified two big questions in one sweep. First, what are the developmental mechanisms underlying the formation and activity of the focus? And second, how does a new organizer evolve, superimposed on the conserved general ground plan of an insect wing?Fred's results had been known for a dozen years, but no molecular studies of the eyespot focus had materialized. This was our golden opportunity because, although we knew very little about butterflies, we were immersed in the genetic dissection of Drosophila wing development. We just used our knowledge from flies to take a look at how butterflies might do things similarly, or better yet, differently. We did, indeed, discover that eyespots were special and formed from unique patterning systems operating within the butterfly wing.Our migration into butterfly wing development opened our eyes to many questions that could be pursued with these beautiful animals and served as our training ground for comparative approaches to more exotic creatures. It has fostered interdisciplinary and collaborative approaches to ecology, genetics, development and evolutionary biology that were not foreseen when we began. Our initial rediscovery of Nijhout's work has led to our rediscovery of classic work on mimicry, melanism, plasticity and other phenomena that beg for a developmental and genetic explanation and that frame our current agenda.Most scientists will admit to occasional serendipitous discoveries. In my case, it's possible that none of this would have happened if I hadn't had the luck to be in the right place (a parking lot in North Carolina) at a time when studies in Drosophila were far enough advanced to allow us to venture off onto roads less traveled.