# Separate keys · fresh Days 65 and 70 checks

Use only after the [first independent response](ASSESSMENT.md) is recorded. These are formative next-move guides, not marks, diagnosis, mastery certificates or claims of a classroom pilot. There is **no fixed real-plant answer**: an actual feature/result needs a school-approved known safe source, local date, comparable viewpoint, exact access and observation. A neutral **ADULT REPORT** supports participation but is not the child's direct sense. No paper/tactile model can establish a local plant feature or actual sensory investigation.

## Day 65 · Case J

- **Strong bounded paper response:** ONE shows four pointed marks; TWO shows three visible marks because the striped flap covers the upper-right corner of the same page. The source says the author has not erased a mark. Therefore the fourth mark is hidden in this paper view, not proven absent from a real plant. Case J supplies no actual plant/date or change.
- **Reasonable next question:** “What external feature is genuinely visible on one school-approved plant from a safe marked place today?” Identify actual source/date/view/access and one feature or UNKNOWN. If a second real view is possible later, record its own date and any changed conditions. Do not force contact or assume a missing feature.
- **Do not award:** “A real leaf fell off”; a real species identification; Day 65 actual AC9SFU01 content or AC9SFI02 safe investigation from J alone; a prior prediction made after seeing the page.
- **Next move:** if hidden/absent is merged, repeat with broad fixed paper flap; if source category is weak, contrast paper/model with a genuine attributed dated school view; if clear, plan a safe second viewpoint without claiming the plant changed.

## Day 70 · Case K

- **Strong bounded paper response:** EARLIER page shows two scalloped marks; LATER page shows three, and the second page is twice print size. The *author labels* both from the mark, but no real plant, actual date or method is supplied. We may compare paper marks and note scale; we cannot say a real leaf appeared or a plant grew. The same-view label is not a verified real observation.
- **Reasonable next check:** record two genuinely approved views of the same real plant from a stable position, on distinct actual local dates, attending to the same visible feature; record source, access and changed conditions. Write a prediction before the second view, then compare with its actual observation if both exist. “Same”, “different” or “unclear” may all be supported by the resulting evidence.
- **Do not award:** a biological growth claim from page size/count; AC9SFI04 actual prediction comparison from fictional pages; a grouping claim from one plant; content mastery of animals. If actual views exist but differ in angle, say INCOMPARABLE or limit the claim.
- **Next move:** if child focuses only on 2/3 marks, ask what source type the pages are; if scale is missed, show two broad fixed cards; if source discipline is strong, make a private cautious comparison of genuine local records.

**Interpretation:** AC9SFU01's full wording covers external features of **plants and animals** and grouping. This pack's genuine plant observations and two-plant grouping, when available, are partial evidence; neither check alone covers the whole row. AC9SFI02 requires an actual safe sensory investigation. AC9SFI04 requires a genuinely earlier prediction and later actual observation. A model-only answer can show source reasoning but never a real plant result. When absent, record **REAL OBSERVATION NOT AVAILABLE** and the next safe opportunity, not inability or measured growth.