**The most counterintuitive finding is that Iris-setosa has the smallest flowers by almost every measure, yet the widest sepals.** This comes from the species-level aggregates I queried (`Iris_by_Species`). - **Sepal width runs against the size ordering.** Setosa's average sepal width is about 3.42 cm, against 2.97 cm for virginica and 2.77 cm for versicolor. - **Every other measure follows the expected small-to-large order.** Setosa is smallest, then versicolor, then virginica. Average sepal length is 5.01, 5.94 and 6.59 cm. Average petal length is 1.46, 4.26 and 5.55 cm. Average petal width is 0.24, 1.33 and 2.03 cm. - **The ordering flips for sepal width, and it is not monotonic.** Setosa is widest, but virginica is wider than versicolor even though it is the larger flower in every other way. "Bigger flower means bigger everything" is false here. Two related points show why this matters for classification: - **Petals separate setosa completely.** Its maximum petal length (1.9 cm) is well below the minimum for versicolor (3.0 cm). Its maximum petal width (0.6 cm) is likewise below versicolor's minimum (1.0 cm). A single petal threshold identifies setosa. - **Sepals separate it poorly.** Setosa's sepal width range (2.3–4.4 cm) overlaps the other two species (versicolor 2.0–3.4, virginica 2.2–3.8). Sepal length overlaps heavily between versicolor (4.9–7.0) and virginica (4.9–7.9). So the feature most people would naturally measure first, sepal size, is a misleading guide to species, while petal dimensions separate the groups much more cleanly. Versicolor and virginica do still overlap on petal length (versicolor up to 5.1 cm, virginica from 4.5 cm). Telling those two apart needs more than a single cut-off.