I happen to know someone who wrote a thing that is publically available as an Open Access article.

Developmental constraint underlies the replicated evolution of grass awns

A five-part figure introducing the awned lemma and mapping awn presence onto a phylogeny.

Awns have been gained, lost, and regained many times in the Pooideae. (a) Generalized grass flower diagram, awn noted with arrow. (b) Diversity of awns in the Pooideae. From top: abaxially insterted twisted geniculate awn (Calamagrostis sp.), apically inserted straight awn (Secale cereale), and apically inserted twisted geniculate awn (Stipa spartea). (c) Predicted homology of awned lemmas to leaves. (d) Ancestral state reconstruction for awn presence. Species of interest noted: (1) brachypodium (Brachypodium distachyon), (2) Bromus tectorum, (3) barley (Hordeum vulgare), (4) rye (Secale cereale), (5) wheat (Triticum aestivum), (6) velvetgrass (Holcus lanatus), (7) blackgrass (Alopecurus myosuroides). Nodes of interest noted: most recent common ancestor of (1) Triticodae, (2) Brachypodium and Triticodae, (3) Poeae and Aveneae. Purple branches indicate gain of awns. Yellow node indicates Pooideae most recent common ancestor. (e) Transition rates between awned and awnless states in (d). Lemma and leaf in (c) from Thi-Tuyet-Hoa (1965), redrawn with permission, copyright Meise Botanic Garden. Images in (b), and (e) from Hitchcock-Chase Collection of Grass Drawings.

https://nph.onlinelibrary.wiley.com/doi/10.1111/nph.20268

I’m pretty proud of the author. Give it a read.