**The most important metric is `petal_length`.** The dataset has no revenue-style target, so I judged importance by how well a metric separates the three species, which is what the dataset is for (choosing classifier features). **Why petal length** - Setosa's petal length runs 1.0–1.9 cm (mean 1.46). That sits completely below versicolor (3.0–5.1, mean 4.26) and virginica (4.5–6.9, mean 5.55), so setosa is perfectly separable on this metric alone. - Versicolor and virginica overlap only in a narrow band, about 4.5–5.1 cm. Their means are also far apart: 4.26 versus 5.55. - Sepal width separates worse. The species ranges overlap heavily (setosa 2.3–4.4, versicolor 2.0–3.4, virginica 2.2–3.8), and versicolor and virginica have nearly the same mean (2.77 versus 2.97). - Petal width behaves almost like petal length (setosa 0.1–0.6, versicolor 1.0–1.8, virginica 1.4–2.5). Petal length is my pick because its between-species gaps are the widest in absolute terms. Petal width is a close second and nearly interchangeable for modelling. **Highest and lowest records by petal length** - **Highest:** a virginica specimen with petal length 6.9 cm, petal width 2.3, sepal length 7.7 and sepal width 2.6. - **Lowest:** a setosa specimen with petal length 1.0 cm, petal width 0.2, sepal length 4.6 and sepal width 3.6. The extremes come from opposite species, which is the same separation described above. The two queries returned one row each, so if several specimens tie at 6.9 or 1.0, only one is shown.