Graph all the things

analyzing all the things you forgot to wonder about



15153565100150200million years agopineapplebananaacaicoconutdragonfruittomatoblueberrykiwigooseberrypomegranatepapayadurianmangomandarinlimelemonorangegrapefruitpomelowatermelonhoneydewcantaloupestrawberryraspberryapplepearcherrypeachplumapricotfloweringplantsmonocotspalmsdicotscitrusrosesprunesmelons15153565100150200million years ago
Fruits selected:

Most recent common ancestor around: 200 million years ago
A biologist might call all these fruits:  "flowering plants" or "angiosperms"

I got curious about the evolutionary history of the fruits we commonly eat. How long ago did they branch off? Which ones are more similar to others? Does genetic similarity correlate with nutritional similarity? (Only slightly.) Are there fruits that diverged through domestication, other than obvious cases like apple varieties? (I couldn't find any.) And are those people who call watermelons "berries" actually basing it on science? (No.)

In the diagram, you can select a subset of fruits to see how long ago they diverged. Sorry for the weird time scale - a linear scale left no detail for the recent past, and a log scale left no detail for the distant past, so I went with a square root scale.

A few fun facts:

  • Honeydew and cantaloupe are very close to cucumber - even closer than they are to watermelon.
  • I never would have guessed that blueberry and kiwi are relatively closely related, but now it occurs to me that they're both green inside.
  • Dragonfruit/pitaya comes from a cactus.
  • A few things are actually a cross: lemon is sour orange x citron. Neither of those ancestors shows up here, but the point is that these tree diagrams hide even more complexity in the evolutionary process.
  • You'd think all kiwis would be related, sharing brown fuzz and being associated with New Zealand, but actually the kiwi fruit and kiwi bird diverged 1.4 billion years ago.
  • One or two of the papers I looked at take a look at the rapid "radiation" of a clade of plants branching off in many ways. You can see one such rapid bifurcation of rosids around 120 million years ago. Perhaps some new evolutionary developments unlocked more biological niches and adaptations around that time.

Minor note about the visualization: I decided to try doing everything in pure React this time rather than d3. It was a bit easier to code, but the responsiveness is lacking since I'm relying on setState. There are ways I could get around this, but I'm pretty sure they're all more complicated than using d3. I find the buttons particularly infurating because of the delay between mouse enter and its reaction. But no, I'm not going to spend time improving it.


I frequently referred to onezoom (which I highly recommend exploring) and Wikipedia when collecting this data. I compiled the chronology data from a few papers:

  • Xiang, Yezi et al. "Evolution of Rosaceae Fruit Types Based on NuclearPhylogeny in the Context of Geological Times andGenome Duplication."
  • Carbonell-Caballero, Jose et al. "A Phylogenetic Analysis of 34 Chloroplast Genomes Elucidates the Relationships between Wild and Domestic Species within the Genus Citrus."
  • Zeng, Liping et al. "Resolution of deep eudicot phylogeny and their temporal diversification using nuclear genes from transcriptomic and genomic datasets."
  • Rose, Jeffrey P. et al. "Phylogeny, historical biogeography, and diversification of angiosperm order Ericales suggest ancient Neotropical and East Asian connections."
  • Ming, Ray et al. "The pineapple genome and the evolution of CAM photosynthesis."

Note that I couldn't get an precise year every single bifurcation. In one or two cases, I couldn't even find a rough number and had to make my own inferences based on the genetic difference data. I originally hoped that finding all the years would be easier, but I'm actually glad it was challenging. Through my struggle, I learned about how these studies work.