The Dissection Table Approach
I spent a lot of time trying to get biology students to properly identify and label floral structures without them mixing up the stamens and carpels, which happens constantly. The standard worksheet method works fine for simple flowers like lilies, but it falls apart fast once you introduce something with whorls of tepals or fused petals, and that is where most people give up. I stopped assigning generic diagrams about five years ago and switched to a hands-on dissection and direct labelling workflow instead. It takes more class time but the retention rate is noticeably better, and honestly it is less work for me to grade. The process starts with a fresh specimen, not a preserved one from a jar, because preserved tissue shrinks and the petal bases curl in ways that obscure the attachment points students need to see. A white or light-coloured flower makes the colour contrast between reproductive parts and vegetative parts clearer, so you will usually want to work with something like a hibiscus, a radish blossom, or a broad bean flower. The broader the petals and the more spaced out the internal structures, the fewer complaints you get about identifying individual parts. You lay the flower face up on a dissection tray, cut away the outermost ring of structures first, and keep moving inward. Each whorl goes onto its own piece of tracing paper or a clean quadrant of the lab sheet so nothing gets lost or stuck to the forceps. The key mistake beginners make is trying to pull the whole flower apart at once, which smashes the delicate stamens against the ovary and makes accurate identification impossible. Work ring by ring, pin each part down as you remove it, then apply the labels directly onto the tracing layer rather than trying to draw lines on the original diagram.
Labelling The Parts Of A Flower: What Goes Where
Here is the actual structure you are dealing with, in the order most field guides and textbooks use. The outermost whorl consists of sepals, collectively called the calyx, which function mainly as a protective sheath while the bud is closed. Inside that is the cornea, made up of petals, which in many species contain pigment cells and scent glands that attract specific pollinators, though in wind-pollinated flowers these structures can be reduced or absent entirely. Moving inward, the androecium is made of stamens, each composed of a filament and an anther. The anther is where you look for the dehisced pollen sacs, usually opening through longitudinal slits called stomia in most common flowering plants, but through pores at the tip in ericaceous species like blueberries, which is a detail that trips people up if they have only ever examined garden flowers. The innermost whorl is the gynoecium, built from carpels. A single carpel contains the stigma, style, and ovary. The stigma receives the pollen, the style is the conduit, and the ovary houses the ovules. In some flowers the carpels are free, meaning each is an independent unit, and in others they are fused into a single pistil structure. That fusion point, where two or more carpels meet, creates a partition called a septum inside the ovary, and determining whether the ovary is superior or inferior based on where the other floral parts attach to it is probably the single most important morphological distinction you will be tested on. A hypogynous flower has the attachment point below the ovary, making it superior, while a perigynous or epigynous flower has the other parts arising from around or above the ovary, making it inferior. Getting this wrong changes the entire botanical description of the specimen. One edge case I ran into repeatedly involves flowers where the petals are actually modified stamens, which is the normal condition in many ornamental cultivars like peonies and certain lilies. Breeding for showier flowers frequently converts stamens into petal-like structures through a process called staminode conversion, so a flower that looks like it has six petals might actually have three true petals and three transformed stamens, or in some cases just three sterile stamens masquerading as petals. When you are labelling a cultivated variety, the standard terminology breaks down unless you know the wild-type ancestral form, which means you sometimes have to fall back on describing what you see rather than applying the textbook label blindly.
Practical Labelling Workflow
After you have isolated each whorl, arrange the parts in order on your lab sheet, starting from the outside and working in. Draw a light circle to represent the floral axis, place the sepals at the periphery, then the petals, stamens, and finally the carpels in the centre. Use fine-point permanent marker for the labels and draw a thin leader line from each term to its corresponding structure. Keep the leader lines uncluttered, because crossing lines make it impossible to verify that your labels match the correct part, and that is exactly what happens when you rush this step. For the ovary, make a vertical cross-section with a fresh razor blade and mount the section on a slide to confirm the number of locules and the placenta type, because the external shape of the ovary does not always reveal the internal ovule arrangement. Paracarpous placentation, where the ovules sit on walls formed by fused carpel edges, looks completely different from axile placentation, where the ovules are on a central column, and you cannot reliably tell the difference without the section. I usually find that showing students the cross-section before they attempt to label the whole flower cuts the number of incorrect ovary descriptions in half.
Get the Full Details

Common Problems and What Actually Helps
The biggest practical issue is that many flowers do not come apart cleanly. Flowers with syncarpous gynoecia, where the carpels are fused, resist separation because the style and stigma often tear when you try to lift the carpel free. The workaround is to cut horizontally through the ovary with a scalpel before attempting any removal, which releases the structural tension and lets the inner whorls lift out intact. This also gives you the ovary cross-section you need for the placentation check at the same time, so you save a step. Another persistent problem involves small flowers where the individual structures are too tiny to handle with standard forceps. A fine brush dipped in water works better than tweezers for moving structures like individual anthers or tiny stigmas under those conditions, and it prevents crushing them. I recommend keeping a soft camel-hair brush in the dissection tray at all times, even if you think the flowers are large enough, because you will always encounter at least one specimen that is smaller than expected.
Verification Without Overcomplicating Things
Once the labels are applied, compare your arrangement against a published floral formula for the species if one is available. A floral formula encodes the number of parts, fusion state, and ovary position in a compact symbolic notation, and it serves as an independent check on whether your identification is consistent with the accepted morphology. For example, a formula like K5 C5 A G(5) tells you the flower has five fused sepals, five free petals, numerous stamens, and a gynoecium of five fused carpels with a superior ovary. If your dissection shows free carpels instead of fused ones, the formula flags the discrepancy immediately. The limitations of this approach are worth noting upfront. It only works reliably on flowers that are accessible and fresh, which excludes many woodland species that bloom briefly and close quickly, as well as any flower with a highly specialized pollination syndrome that causes rapid post-pollination structural changes. In those cases, relying on herbarium specimens or photographic keys is more practical, though herbarium material lacks the turgidity needed for clean dissection. There is no perfect workaround for that trade-off, so you choose the method based on what material is actually available rather than pretending one method covers everything.