P. 354 of Theorem List - Metamath Proof Explorer

Theorem List for Metamath Proof Explorer - 35301-35400   *Has distinct variable group(s)

Type	Label	Description
Statement
Theorem	r1ssel 35301	A set is a subset of the value of the cumulative hierarchy of sets function iff it is an element of the value at the successor. (Contributed by BTernaryTau, 15-Jan-2026.)
⊢ (𝐵 ∈ On → (𝐴 ⊆ (𝑅1‘𝐵) ↔ 𝐴 ∈ (𝑅1‘suc 𝐵)))
Theorem	axnulALT3 35302*	Alternate proof of axnul 5229, proved from propositional calculus, ax-gen 1803, ax-4 1817, ax-5 1918, and ax-inf2 9557. (Contributed by BTernaryTau, 22-Jun-2025.) (Proof modification is discouraged.) (New usage is discouraged.)
⊢ ∃𝑥∀𝑦 ¬ 𝑦 ∈ 𝑥
Theorem	axprALT2 35303*	Alternate proof of axpr 5358, proved from predicate calculus, ax-rep 5201, and ax-inf2 9557. (Contributed by BTernaryTau, 26-Mar-2026.) (Proof modification is discouraged.) (New usage is discouraged.)
⊢ ∃𝑧∀𝑤((𝑤 = 𝑥 ∨ 𝑤 = 𝑦) → 𝑤 ∈ 𝑧)
Theorem	r1omfv 35304	Value of the cumulative hierarchy of sets function at ω. (Contributed by BTernaryTau, 25-Jan-2026.)
⊢ (𝑅1‘ω) = ∪ (𝑅1 “ ω)
Theorem	trssfir1om 35305	If every element in a transitive class is finite, then every element is also hereditarily finite. (Contributed by BTernaryTau, 24-Jan-2026.)
⊢ ((Tr 𝐴 ∧ 𝐴 ⊆ Fin) → 𝐴 ⊆ ∪ (𝑅1 “ ω))
Theorem	r1omhfb 35306*	The class of all hereditarily finite sets is the only class with the property that all sets are members of it iff they are finite and all of their elements are members of it. (Contributed by BTernaryTau, 24-Jan-2026.)
⊢ (𝐻 = ∪ (𝑅1 “ ω) ↔ ∀𝑥(𝑥 ∈ 𝐻 ↔ (𝑥 ∈ Fin ∧ ∀𝑦 ∈ 𝑥 𝑦 ∈ 𝐻)))
21.5.2.1  Finitism
Theorem	prcinf 35307*	Any proper class is literally infinite, in the sense that it contains subsets of arbitrarily large finite cardinality. This proof holds regardless of whether the Axiom of Infinity is accepted or negated. (Contributed by BTernaryTau, 22-Jun-2025.)
⊢ (¬ 𝐴 ∈ V → ∀𝑛 ∈ ω ∃𝑥(𝑥 ⊆ 𝐴 ∧ 𝑥 ≈ 𝑛))
Theorem	fineqvrep 35308*	If all sets are finite, then the Axiom of Replacement becomes redundant. (Contributed by BTernaryTau, 12-Sep-2024.)
⊢ (Fin = V → (∀𝑤∃𝑦∀𝑧(∀𝑦𝜑 → 𝑧 = 𝑦) → ∃𝑦∀𝑧(𝑧 ∈ 𝑦 ↔ ∃𝑤(𝑤 ∈ 𝑥 ∧ ∀𝑦𝜑))))
Theorem	fineqvpow 35309*	If all sets are finite, then the Axiom of Power Sets becomes redundant. (Contributed by BTernaryTau, 12-Sep-2024.)
⊢ (Fin = V → ∃𝑦∀𝑧(∀𝑤(𝑤 ∈ 𝑧 → 𝑤 ∈ 𝑥) → 𝑧 ∈ 𝑦))
Theorem	fineqvac 35310	If all sets are finite, then the Axiom of Choice becomes redundant. For a shorter proof using ax-rep 5201 and ax-pow 5296, see fineqvacALT 35311. (Contributed by BTernaryTau, 21-Sep-2024.)
⊢ (Fin = V → CHOICE)
Theorem	fineqvacALT 35311	Shorter proof of fineqvac 35310 using ax-rep 5201 and ax-pow 5296. (Contributed by BTernaryTau, 21-Sep-2024.) (Proof modification is discouraged.) (New usage is discouraged.)
⊢ (Fin = V → CHOICE)
Theorem	fineqvomon 35312	If all sets are finite, then the class of all natural numbers equals the proper class of all ordinal numbers. (Contributed by BTernaryTau, 30-Dec-2025.)
⊢ (Fin = V → ω = On)
Theorem	fineqvomonb 35313	All sets are finite iff all ordinal sets are finite. (Contributed by BTernaryTau, 25-Jan-2026.)
⊢ (Fin = V ↔ ω = On)
Theorem	omprcomonb 35314	The class of all finite ordinals is a proper class iff all ordinal sets are finite. (Contributed by BTernaryTau, 25-Jan-2026.)
⊢ (¬ ω ∈ V ↔ ω = On)
Theorem	fineqvnttrclselem1 35315*	Lemma for fineqvnttrclse 35318. (Contributed by BTernaryTau, 12-Jan-2026.)
⊢ (𝐵 ∈ (ω ∖ 1o) → ∪ {𝑑 ∈ On ∣ (𝐴 +o 𝑑) = 𝐵} ∈ ω)
Theorem	fineqvnttrclselem2 35316*	Lemma for fineqvnttrclse 35318. (Contributed by BTernaryTau, 12-Jan-2026.)
⊢ 𝐹 = (𝑣 ∈ suc suc 𝑁 ↦ ∪ {𝑑 ∈ On ∣ (𝑣 +o 𝑑) = 𝐵})    ⇒   ⊢ ((𝐵 ∈ (ω ∖ 1o) ∧ 𝑁 ∈ 𝐵 ∧ 𝐴 ∈ suc suc 𝑁) → (𝐴 +o (𝐹‘𝐴)) = 𝐵)
Theorem	fineqvnttrclselem3 35317*	Lemma for fineqvnttrclse 35318. (Contributed by BTernaryTau, 12-Jan-2026.)
⊢ 𝑅 = {⟨𝑥, 𝑦⟩ ∣ (𝑥 ∈ 𝐴 ∧ 𝑥 = suc 𝑦)}    &   ⊢ 𝐴 = ω    &   ⊢ 𝐹 = (𝑣 ∈ suc suc 𝑁 ↦ ∪ {𝑑 ∈ On ∣ (𝑣 +o 𝑑) = 𝐵})    ⇒   ⊢ ((𝐵 ∈ (ω ∖ 1o) ∧ 𝑁 ∈ 𝐵) → ∀𝑎 ∈ suc 𝑁(𝐹‘𝑎)𝑅(𝐹‘suc 𝑎))
Theorem	fineqvnttrclse 35318*	A counterexample demonstrating that ttrclse 9643 does not hold when all sets are finite. (Contributed by BTernaryTau, 12-Jan-2026.)
⊢ 𝑅 = {⟨𝑥, 𝑦⟩ ∣ (𝑥 ∈ 𝐴 ∧ 𝑥 = suc 𝑦)}    &   ⊢ 𝐴 = ω    ⇒   ⊢ (Fin = V → (𝑅 Se 𝐴 ∧ ¬ t++(𝑅 ↾ 𝐴) Se 𝐴))
Theorem	fineqvinfep 35319*	A counterexample demonstrating that tz9.1 9645 does not hold when all sets are finite and an infinite descending ∈-chain exists. (Contributed by BTernaryTau, 18-Feb-2026.)
⊢ 𝐴 = {(𝐹‘∅)}    ⇒   ⊢ ((Fin = V ∧ 𝐹:ω–1-1→V ∧ ∀𝑥 ∈ ω (𝐹‘suc 𝑥) ∈ (𝐹‘𝑥)) → ¬ ∃𝑦(𝐴 ⊆ 𝑦 ∧ Tr 𝑦))
21.5.2.2  Introduce ax-regs
Axiom	ax-regs 35320*	A strong version of the Axiom of Regularity. It states that if there exists a set with property 𝜑, then there must exist a set with property 𝜑 such that none of its elements have property 𝜑. This axiom can be derived from the axioms of ZF set theory as shown in axregs 35333, but this derivation relies on ax-inf2 9557 and is thus not possible in a finitist context. (Contributed by BTernaryTau, 29-Dec-2025.)
⊢ (∃𝑥𝜑 → ∃𝑦(∀𝑥(𝑥 = 𝑦 → 𝜑) ∧ ∀𝑧(𝑧 ∈ 𝑦 → ¬ ∀𝑥(𝑥 = 𝑧 → 𝜑))))
Theorem	axreg 35321*	Derivation of ax-reg 9501 from ax-regs 35320 and Tarski's FOL axiom schemes. This demonstrates the sense in which ax-regs 35320 is a stronger version of ax-reg 9501. (Contributed by BTernaryTau, 30-Dec-2025.)
⊢ (∃𝑦 𝑦 ∈ 𝑥 → ∃𝑦(𝑦 ∈ 𝑥 ∧ ∀𝑧(𝑧 ∈ 𝑦 → ¬ 𝑧 ∈ 𝑥)))
Theorem	axregscl 35322*	A version of ax-regs 35320 with a class variable instead of a wff variable. Axiom D in Gödel, The Consistency of the Axiom of Choice and of the Generalized Continuum Hypothesis with the Axioms of Set Theory (1940), p. 6. (Contributed by BTernaryTau, 30-Dec-2025.)
⊢ (∃𝑥 𝑥 ∈ 𝐴 → ∃𝑦(𝑦 ∈ 𝐴 ∧ ∀𝑧(𝑧 ∈ 𝑦 → ¬ 𝑧 ∈ 𝐴)))
Theorem	axregszf 35323*	Derivation of zfregs 9648 using ax-regs 35320. (Contributed by BTernaryTau, 30-Dec-2025.)
⊢ (𝐴 ≠ ∅ → ∃𝑥 ∈ 𝐴 (𝑥 ∩ 𝐴) = ∅)
Theorem	setindregs 35324*	Set (epsilon) induction. This version of setind 9663 replaces zfregs 9648 with axregszf 35323. (Contributed by BTernaryTau, 30-Dec-2025.)
⊢ (∀𝑥(𝑥 ⊆ 𝐴 → 𝑥 ∈ 𝐴) → 𝐴 = V)
Theorem	setinds2regs 35325*	Principle of set induction (or E-induction). If a property passes from all elements of 𝑥 to 𝑥 itself, then it holds for all 𝑥. (Contributed by BTernaryTau, 31-Dec-2025.)
⊢ (𝑥 = 𝑦 → (𝜑 ↔ 𝜓))    &   ⊢ (∀𝑦 ∈ 𝑥 𝜓 → 𝜑)    ⇒   ⊢ 𝜑
Theorem	noinfepfnregs 35326*	There are no infinite descending ∈-chains, proven using ax-regs 35320. (Contributed by BTernaryTau, 18-Feb-2026.)
⊢ (𝐹 Fn ω → ∃𝑥 ∈ ω (𝐹‘suc 𝑥) ∉ (𝐹‘𝑥))
Theorem	noinfepregs 35327*	There are no infinite descending ∈-chains, proven using ax-regs 35320. (Contributed by BTernaryTau, 18-Feb-2026.)
⊢ ∃𝑥 ∈ ω (𝐹‘suc 𝑥) ∉ (𝐹‘𝑥)
Theorem	tz9.1regs 35328*	Every set has a transitive closure (the smallest transitive extension). This version of tz9.1 9645 depends on ax-regs 35320 instead of ax-reg 9501 and ax-inf2 9557. This suggests a possible answer to the third question posed in tz9.1 9645, namely that the missing property is that countably infinite classes must obey regularity. In ZF set theory we can prove this by showing that countably infinite classes are sets and thus ax-reg 9501 applies to them directly, but in a finitist context it seems that an axiom like ax-regs 35320 is required since countably infinite classes are proper classes. A related candidate for the missing property is the non-existence of infinite descending ∈-chains, proven as noinfep 9576 using ax-reg 9501 and ax-inf2 9557 and as noinfepregs 35327 using ax-regs 35320. If all sets are finite, then the existence of such a chain implies there is a set which does not have a transitive closure, as shown in fineqvinfep 35319. (Contributed by BTernaryTau, 31-Dec-2025.)
⊢ 𝐴 ∈ V    ⇒   ⊢ ∃𝑥(𝐴 ⊆ 𝑥 ∧ Tr 𝑥 ∧ ∀𝑦((𝐴 ⊆ 𝑦 ∧ Tr 𝑦) → 𝑥 ⊆ 𝑦))
Theorem	unir1regs 35329	The cumulative hierarchy of sets covers the universe. This version of unir1 9732 replaces setind 9663 with setindregs 35324. (Contributed by BTernaryTau, 30-Dec-2025.)
⊢ ∪ (𝑅1 “ On) = V
Theorem	trssfir1omregs 35330	If every element in a transitive class is finite, then every element is also hereditarily finite. This version of trssfir1om 35305 replaces setinds2 9667 with setinds2regs 35325. (Contributed by BTernaryTau, 20-Jan-2026.)
⊢ ((Tr 𝐴 ∧ 𝐴 ⊆ Fin) → 𝐴 ⊆ ∪ (𝑅1 “ ω))
Theorem	r1omhfbregs 35331*	The class of all hereditarily finite sets is the only class with the property that all sets are members of it iff they are finite and all of their elements are members of it. This version of r1omhfb 35306 replaces setinds2 9667 with setinds2regs 35325 and trssfir1om 35305 with trssfir1omregs 35330. (Contributed by BTernaryTau, 21-Jan-2026.)
⊢ (𝐻 = ∪ (𝑅1 “ ω) ↔ ∀𝑥(𝑥 ∈ 𝐻 ↔ (𝑥 ∈ Fin ∧ ∀𝑦 ∈ 𝑥 𝑦 ∈ 𝐻)))
Theorem	fineqvr1ombregs 35332	All sets are finite iff all sets are hereditarily finite. (Contributed by BTernaryTau, 30-Dec-2025.)
⊢ (Fin = V ↔ ∪ (𝑅1 “ ω) = V)
21.5.2.3  Derive ax-regs
Theorem	axregs 35333*	Derivation of ax-regs 35320 from the axioms of ZF set theory. (Contributed by BTernaryTau, 29-Dec-2025.)
⊢ (∃𝑥𝜑 → ∃𝑦(∀𝑥(𝑥 = 𝑦 → 𝜑) ∧ ∀𝑧(𝑧 ∈ 𝑦 → ¬ ∀𝑥(𝑥 = 𝑧 → 𝜑))))
21.5.2.4  ZFC axioms with reduced distinct variable conditions
Theorem	axsepg2 35334*	A generalization of ax-sep 5220 in which 𝑥 and 𝑧 need not be distinct. This theorem scheme bundles ax-sep 5220 with the degenerate instance ∃𝑦∀𝑧(𝑧 ∈ 𝑦 ↔ (𝑧 ∈ 𝑧 ∧ 𝜑)) which is satisfied by the existence of the empty set. Usage of this theorem is discouraged because it depends on ax-13 2382. (Contributed by BTernaryTau, 21-May-2026.) (New usage is discouraged.)
⊢ ∃𝑦∀𝑥(𝑥 ∈ 𝑦 ↔ (𝑥 ∈ 𝑧 ∧ 𝜑))
Theorem	axsepg3 35335*	A generalization of ax-sep 5220 in which 𝑦 and 𝑧 need not be distinct. This theorem scheme bundles ax-sep 5220 with the degenerate instance ∃𝑦∀𝑥(𝑥 ∈ 𝑦 ↔ (𝑥 ∈ 𝑦 ∧ 𝜑)) which is satisfied by the existence of the empty set. Usage of this theorem is discouraged because it depends on ax-13 2382. (Contributed by BTernaryTau, 3-Aug-2025.) (New usage is discouraged.)
⊢ ∃𝑦∀𝑥(𝑥 ∈ 𝑦 ↔ (𝑥 ∈ 𝑧 ∧ 𝜑))
Theorem	axsepg3ALT 35336*	Alternate proof of axsepg3 35335, derived directly from ax-sep 5220 with no additional set theory axioms. (Contributed by BTernaryTau, 3-Aug-2025.) (Proof modification is discouraged.) (New usage is discouraged.)
⊢ ∃𝑦∀𝑥(𝑥 ∈ 𝑦 ↔ (𝑥 ∈ 𝑧 ∧ 𝜑))
Theorem	axsepg4 35337*	A generalization of ax-sep 5220 that combines axsepg 5221 and axsepg2 35334 into a single theorem scheme. Unlike ax-sep 5220, this scheme lacks a distinct variable condition for 𝜑 and 𝑧 as well as for 𝑥 and 𝑧. Usage of this theorem is discouraged because it depends on ax-13 2382. (Contributed by BTernaryTau, 24-May-2026.) (New usage is discouraged.)
⊢ ∃𝑦∀𝑥(𝑥 ∈ 𝑦 ↔ (𝑥 ∈ 𝑧 ∧ 𝜑))
Theorem	axsepg5 35338*	A generalization of ax-sep 5220 that combines axsepg 5221, axsepg2 35334, and axsepg3 35335 into a single theorem scheme. Unlike ax-sep 5220, this scheme lacks a distinct variable condition for 𝜑 and 𝑧, for 𝑥 and 𝑧, and for 𝑦 and 𝑧. Usage of this theorem is discouraged because it depends on ax-13 2382. (Contributed by BTernaryTau, 24-May-2026.) (New usage is discouraged.)
⊢ ∃𝑦∀𝑥(𝑥 ∈ 𝑦 ↔ (𝑥 ∈ 𝑧 ∧ 𝜑))
Theorem	axnulg 35339	A generalization of ax-nul 5230 in which 𝑥 and 𝑦 need not be distinct. This theorem scheme bundles ax-nul 5230 with the degenerate instance ∃𝑥∀𝑥¬ 𝑥 ∈ 𝑥 which is satisfied by elirrv 9506. Usage of this theorem is discouraged because it depends on ax-13 2382. (Contributed by BTernaryTau, 3-Aug-2025.) (New usage is discouraged.)
⊢ ∃𝑥∀𝑦 ¬ 𝑦 ∈ 𝑥
Theorem	axpowg 35340*	A generalization of ax-pow 5296 that combines it and zfpow 5297 into a single theorem scheme. Unlike ax-pow 5296, this scheme lacks a distinct variable condition for 𝑦 and 𝑤. (Contributed by BTernaryTau, 26-May-2026.)
⊢ ∃𝑦∀𝑧(∀𝑤(𝑤 ∈ 𝑧 → 𝑤 ∈ 𝑥) → 𝑧 ∈ 𝑦)
Theorem	axpowg2 35341*	A generalization of ax-pow 5296 in which 𝑥 and 𝑤 need not be distinct. This theorem scheme bundles ax-pow 5296 with the degenerate instance ∃𝑦∀𝑧(∀𝑥(𝑥 ∈ 𝑧 → 𝑥 ∈ 𝑥) → 𝑧 ∈ 𝑦) which is satisfied by the existence of a set that contains all empty sets (see axprlem1 5354). Usage of this theorem is discouraged because it depends on ax-13 2382. (Contributed by BTernaryTau, 26-May-2026.) (New usage is discouraged.)
⊢ ∃𝑦∀𝑧(∀𝑤(𝑤 ∈ 𝑧 → 𝑤 ∈ 𝑥) → 𝑧 ∈ 𝑦)
Theorem	axpowg3 35342*	A generalization of ax-pow 5296 that combines axpowg 35340 and axpowg2 35341 into a single theorem scheme. Unlike ax-pow 5296, this scheme lacks a distinct variable condition for 𝑦 and 𝑤 as well as for 𝑥 and 𝑤. Usage of this theorem is discouraged because it depends on ax-13 2382. (Contributed by BTernaryTau, 26-May-2026.) (New usage is discouraged.)
⊢ ∃𝑦∀𝑧(∀𝑤(𝑤 ∈ 𝑧 → 𝑤 ∈ 𝑥) → 𝑧 ∈ 𝑦)
21.5.2.5  Global choice
Theorem	gblacfnacd 35343*	If 𝐺 is a global choice function, then the Axiom of Choice (in the form of the right-hand side of dfac4 10039) holds. Note that 𝐺 must be a proper class by fndmexb 7850. This means we cannot show that the existence of a class that behaves as a global choice function is sufficient because we only have existential quantifiers for sets, not (proper) classes. However, if a class variant of exlimiv 1938 were available, then it could be used alongside the closed form of this theorem to prove that result. (Contributed by BTernaryTau, 12-Dec-2024.)
⊢ (𝜑 → 𝐺 Fn V)    &   ⊢ (𝜑 → ∀𝑧(𝑧 ≠ ∅ → (𝐺‘𝑧) ∈ 𝑧))    ⇒   ⊢ (𝜑 → ∀𝑥∃𝑓(𝑓 Fn 𝑥 ∧ ∀𝑧 ∈ 𝑥 (𝑧 ≠ ∅ → (𝑓‘𝑧) ∈ 𝑧)))
Theorem	onvf1odlem1 35344*	Lemma for onvf1od 35348. (Contributed by BTernaryTau, 2-Dec-2025.)
⊢ (𝐴 ∈ 𝑉 → ∃𝑥 ∈ On ∃𝑦 ∈ (𝑅1‘𝑥) ¬ 𝑦 ∈ 𝐴)
Theorem	onvf1odlem2 35345*	Lemma for onvf1od 35348. (Contributed by BTernaryTau, 2-Dec-2025.)
⊢ (𝜑 → ∀𝑧(𝑧 ≠ ∅ → (𝐺‘𝑧) ∈ 𝑧))    &   ⊢ 𝑀 = ∩ {𝑥 ∈ On ∣ ∃𝑦 ∈ (𝑅1‘𝑥) ¬ 𝑦 ∈ 𝐴}    &   ⊢ 𝑁 = (𝐺‘((𝑅1‘𝑀) ∖ 𝐴))    ⇒   ⊢ (𝜑 → (𝐴 ∈ 𝑉 → 𝑁 ∈ ((𝑅1‘𝑀) ∖ 𝐴)))
Theorem	onvf1odlem3 35346*	Lemma for onvf1od 35348. The value of 𝐹 at an ordinal 𝐴. (Contributed by BTernaryTau, 2-Dec-2025.)
⊢ 𝑀 = ∩ {𝑥 ∈ On ∣ ∃𝑦 ∈ (𝑅1‘𝑥) ¬ 𝑦 ∈ ran 𝑤}    &   ⊢ 𝑁 = (𝐺‘((𝑅1‘𝑀) ∖ ran 𝑤))    &   ⊢ 𝐹 = recs((𝑤 ∈ V ↦ 𝑁))    &   ⊢ 𝐵 = ∩ {𝑢 ∈ On ∣ ∃𝑣 ∈ (𝑅1‘𝑢) ¬ 𝑣 ∈ (𝐹 “ 𝐴)}    &   ⊢ 𝐶 = (𝐺‘((𝑅1‘𝐵) ∖ (𝐹 “ 𝐴)))    ⇒   ⊢ (𝐴 ∈ On → (𝐹‘𝐴) = 𝐶)
Theorem	onvf1odlem4 35347*	Lemma for onvf1od 35348. If the range of 𝐹 does not exist, then it must equal the universe. (Contributed by BTernaryTau, 4-Dec-2025.)
⊢ (𝜑 → ∀𝑧(𝑧 ≠ ∅ → (𝐺‘𝑧) ∈ 𝑧))    &   ⊢ 𝑀 = ∩ {𝑥 ∈ On ∣ ∃𝑦 ∈ (𝑅1‘𝑥) ¬ 𝑦 ∈ ran 𝑤}    &   ⊢ 𝑁 = (𝐺‘((𝑅1‘𝑀) ∖ ran 𝑤))    &   ⊢ 𝐹 = recs((𝑤 ∈ V ↦ 𝑁))    &   ⊢ 𝐵 = ∩ {𝑢 ∈ On ∣ ∃𝑣 ∈ (𝑅1‘𝑢) ¬ 𝑣 ∈ (𝐹 “ 𝑡)}    &   ⊢ 𝐶 = (𝐺‘((𝑅1‘𝐵) ∖ (𝐹 “ 𝑡)))    ⇒   ⊢ (𝜑 → (¬ ran 𝐹 ∈ V → ran 𝐹 = V))
Theorem	onvf1od 35348*	If 𝐺 is a global choice function, then 𝐹 is a bijection from the ordinals to the universe. This is the ZFC version of (1 → 2) in https://tinyurl.com/hamkins-gblac. (Contributed by BTernaryTau, 5-Dec-2025.)
⊢ (𝜑 → ∀𝑧(𝑧 ≠ ∅ → (𝐺‘𝑧) ∈ 𝑧))    &   ⊢ 𝑀 = ∩ {𝑥 ∈ On ∣ ∃𝑦 ∈ (𝑅1‘𝑥) ¬ 𝑦 ∈ ran 𝑤}    &   ⊢ 𝑁 = (𝐺‘((𝑅1‘𝑀) ∖ ran 𝑤))    &   ⊢ 𝐹 = recs((𝑤 ∈ V ↦ 𝑁))    ⇒   ⊢ (𝜑 → 𝐹:On–1-1-onto→V)
Theorem	vonf1owev 35349*	If 𝐹 is a bijection from the universe to the ordinals, then 𝑅 well-orders the universe. This is the ZFC version of (2 → 3) in https://tinyurl.com/hamkins-gblac. (Contributed by BTernaryTau, 6-Dec-2025.)
⊢ 𝑅 = {⟨𝑥, 𝑦⟩ ∣ (𝐹‘𝑥) ∈ (𝐹‘𝑦)}    ⇒   ⊢ (𝐹:V–1-1-onto→On → 𝑅 We V)
Theorem	wevgblacfn 35350*	If 𝑅 is a well-ordering of the universe, then 𝐺 is a global choice function. Here 𝐺 maps each set 𝑧 to its minimal element with respect to 𝑅 (except when 𝑧 is the empty set, in which case it is mapped to the empty set, though this is only done for convenience). This is the ZFC version of (3 → 1) in https://tinyurl.com/hamkins-gblac. (Contributed by BTernaryTau, 29-Jun-2025.)
⊢ 𝐺 = (𝑧 ∈ V ↦ ∪ {𝑦 ∈ 𝑧 ∣ ∀𝑥 ∈ 𝑧 ¬ 𝑥𝑅𝑦})    ⇒   ⊢ (𝑅 We V → (𝐺 Fn V ∧ ∀𝑧(𝑧 ≠ ∅ → (𝐺‘𝑧) ∈ 𝑧)))
21.5.3  Real and complex numbers
Theorem	zltp1ne 35351	Integer ordering relation. (Contributed by BTernaryTau, 24-Sep-2023.)
⊢ ((𝐴 ∈ ℤ ∧ 𝐵 ∈ ℤ) → ((𝐴 + 1) < 𝐵 ↔ (𝐴 < 𝐵 ∧ 𝐵 ≠ (𝐴 + 1))))
Theorem	nnltp1ne 35352	Positive integer ordering relation. (Contributed by BTernaryTau, 24-Sep-2023.)
⊢ ((𝐴 ∈ ℕ ∧ 𝐵 ∈ ℕ) → ((𝐴 + 1) < 𝐵 ↔ (𝐴 < 𝐵 ∧ 𝐵 ≠ (𝐴 + 1))))
Theorem	nn0ltp1ne 35353	Nonnegative integer ordering relation. (Contributed by BTernaryTau, 24-Sep-2023.)
⊢ ((𝐴 ∈ ℕ0 ∧ 𝐵 ∈ ℕ0) → ((𝐴 + 1) < 𝐵 ↔ (𝐴 < 𝐵 ∧ 𝐵 ≠ (𝐴 + 1))))
Theorem	0nn0m1nnn0 35354	A number is zero if and only if it's a nonnegative integer that becomes negative after subtracting 1. (Contributed by BTernaryTau, 30-Sep-2023.)
⊢ (𝑁 = 0 ↔ (𝑁 ∈ ℕ0 ∧ ¬ (𝑁 − 1) ∈ ℕ0))
Theorem	f1resfz0f1d 35355	If a function with a sequence of nonnegative integers (starting at 0) as its domain is one-to-one when 0 is removed, and if the range of that restriction does not contain the function's value at the removed integer, then the function is itself one-to-one. (Contributed by BTernaryTau, 4-Oct-2023.)
⊢ (𝜑 → 𝐾 ∈ ℕ0)    &   ⊢ (𝜑 → 𝐹:(0...𝐾)⟶𝑉)    &   ⊢ (𝜑 → (𝐹 ↾ (1...𝐾)):(1...𝐾)–1-1→𝑉)    &   ⊢ (𝜑 → ((𝐹 “ {0}) ∩ (𝐹 “ (1...𝐾))) = ∅)    ⇒   ⊢ (𝜑 → 𝐹:(0...𝐾)–1-1→𝑉)
Theorem	fisshasheq 35356	A finite set is equal to its subset if they are the same size. (Contributed by BTernaryTau, 3-Oct-2023.)
⊢ ((𝐵 ∈ Fin ∧ 𝐴 ⊆ 𝐵 ∧ (♯‘𝐴) = (♯‘𝐵)) → 𝐴 = 𝐵)
Theorem	revpfxsfxrev 35357	The reverse of a prefix of a word is equal to the same-length suffix of the reverse of that word. (Contributed by BTernaryTau, 2-Dec-2023.)
⊢ ((𝑊 ∈ Word 𝑉 ∧ 𝐿 ∈ (0...(♯‘𝑊))) → (reverse‘(𝑊 prefix 𝐿)) = ((reverse‘𝑊) substr ⟨((♯‘𝑊) − 𝐿), (♯‘𝑊)⟩))
Theorem	swrdrevpfx 35358	A subword expressed in terms of reverses and prefixes. (Contributed by BTernaryTau, 3-Dec-2023.)
⊢ ((𝑊 ∈ Word 𝑉 ∧ 𝐹 ∈ (0...𝐿) ∧ 𝐿 ∈ (0...(♯‘𝑊))) → (𝑊 substr ⟨𝐹, 𝐿⟩) = (reverse‘((reverse‘(𝑊 prefix 𝐿)) prefix (𝐿 − 𝐹))))
21.5.4  Graph theory
Theorem	lfuhgr 35359*	A hypergraph is loop-free if and only if every edge connects at least two vertices. (Contributed by BTernaryTau, 15-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐼 = (iEdg‘𝐺)    ⇒   ⊢ (𝐺 ∈ UHGraph → (𝐼:dom 𝐼⟶{𝑥 ∈ 𝒫 𝑉 ∣ 2 ≤ (♯‘𝑥)} ↔ ∀𝑥 ∈ (Edg‘𝐺)2 ≤ (♯‘𝑥)))
Theorem	lfuhgr2 35360*	A hypergraph is loop-free if and only if every edge is not a loop. (Contributed by BTernaryTau, 15-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐼 = (iEdg‘𝐺)    ⇒   ⊢ (𝐺 ∈ UHGraph → (𝐼:dom 𝐼⟶{𝑥 ∈ 𝒫 𝑉 ∣ 2 ≤ (♯‘𝑥)} ↔ ∀𝑥 ∈ (Edg‘𝐺)(♯‘𝑥) ≠ 1))
Theorem	lfuhgr3 35361*	A hypergraph is loop-free if and only if none of its edges connect to only one vertex. (Contributed by BTernaryTau, 15-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐼 = (iEdg‘𝐺)    ⇒   ⊢ (𝐺 ∈ UHGraph → (𝐼:dom 𝐼⟶{𝑥 ∈ 𝒫 𝑉 ∣ 2 ≤ (♯‘𝑥)} ↔ ¬ ∃𝑎{𝑎} ∈ (Edg‘𝐺)))
Theorem	cplgredgex 35362*	Any two (distinct) vertices in a complete graph are connected to each other by at least one edge. (Contributed by BTernaryTau, 2-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐸 = (Edg‘𝐺)    ⇒   ⊢ (𝐺 ∈ ComplGraph → ((𝐴 ∈ 𝑉 ∧ 𝐵 ∈ (𝑉 ∖ {𝐴})) → ∃𝑒 ∈ 𝐸 {𝐴, 𝐵} ⊆ 𝑒))
Theorem	cusgredgex 35363	Any two (distinct) vertices in a complete simple graph are connected to each other by an edge. (Contributed by BTernaryTau, 3-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐸 = (Edg‘𝐺)    ⇒   ⊢ (𝐺 ∈ ComplUSGraph → ((𝐴 ∈ 𝑉 ∧ 𝐵 ∈ (𝑉 ∖ {𝐴})) → {𝐴, 𝐵} ∈ 𝐸))
Theorem	cusgredgex2 35364	Any two distinct vertices in a complete simple graph are connected to each other by an edge. (Contributed by BTernaryTau, 4-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐸 = (Edg‘𝐺)    ⇒   ⊢ (𝐺 ∈ ComplUSGraph → ((𝐴 ∈ 𝑉 ∧ 𝐵 ∈ 𝑉 ∧ 𝐴 ≠ 𝐵) → {𝐴, 𝐵} ∈ 𝐸))
Theorem	pfxwlk 35365	A prefix of a walk is a walk. (Contributed by BTernaryTau, 2-Dec-2023.)
⊢ ((𝐹(Walks‘𝐺)𝑃 ∧ 𝐿 ∈ (0...(♯‘𝐹))) → (𝐹 prefix 𝐿)(Walks‘𝐺)(𝑃 prefix (𝐿 + 1)))
Theorem	revwlk 35366	The reverse of a walk is a walk. (Contributed by BTernaryTau, 30-Nov-2023.)
⊢ (𝐹(Walks‘𝐺)𝑃 → (reverse‘𝐹)(Walks‘𝐺)(reverse‘𝑃))
Theorem	revwlkb 35367	Two words represent a walk if and only if their reverses also represent a walk. (Contributed by BTernaryTau, 4-Dec-2023.)
⊢ ((𝐹 ∈ Word 𝑊 ∧ 𝑃 ∈ Word 𝑈) → (𝐹(Walks‘𝐺)𝑃 ↔ (reverse‘𝐹)(Walks‘𝐺)(reverse‘𝑃)))
Theorem	swrdwlk 35368	Two matching subwords of a walk also represent a walk. (Contributed by BTernaryTau, 7-Dec-2023.)
⊢ ((𝐹(Walks‘𝐺)𝑃 ∧ 𝐵 ∈ (0...𝐿) ∧ 𝐿 ∈ (0...(♯‘𝐹))) → (𝐹 substr ⟨𝐵, 𝐿⟩)(Walks‘𝐺)(𝑃 substr ⟨𝐵, (𝐿 + 1)⟩))
Theorem	pthhashvtx 35369	A graph containing a path has at least as many vertices as there are edges in the path. (Contributed by BTernaryTau, 5-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    ⇒   ⊢ (𝐹(Paths‘𝐺)𝑃 → (♯‘𝐹) ≤ (♯‘𝑉))
Theorem	spthcycl 35370	A walk is a trivial path if and only if it is both a simple path and a cycle. (Contributed by BTernaryTau, 8-Oct-2023.)
⊢ ((𝐹(Paths‘𝐺)𝑃 ∧ 𝐹 = ∅) ↔ (𝐹(SPaths‘𝐺)𝑃 ∧ 𝐹(Cycles‘𝐺)𝑃))
Theorem	usgrgt2cycl 35371	A non-trivial cycle in a simple graph has a length greater than 2. (Contributed by BTernaryTau, 24-Sep-2023.)
⊢ ((𝐺 ∈ USGraph ∧ 𝐹(Cycles‘𝐺)𝑃 ∧ 𝐹 ≠ ∅) → 2 < (♯‘𝐹))
Theorem	usgrcyclgt2v 35372	A simple graph with a non-trivial cycle must have at least 3 vertices. (Contributed by BTernaryTau, 5-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    ⇒   ⊢ ((𝐺 ∈ USGraph ∧ 𝐹(Cycles‘𝐺)𝑃 ∧ 𝐹 ≠ ∅) → 2 < (♯‘𝑉))
Theorem	subgrwlk 35373	If a walk exists in a subgraph of a graph 𝐺, then that walk also exists in 𝐺. (Contributed by BTernaryTau, 22-Oct-2023.)
⊢ (𝑆 SubGraph 𝐺 → (𝐹(Walks‘𝑆)𝑃 → 𝐹(Walks‘𝐺)𝑃))
Theorem	subgrtrl 35374	If a trail exists in a subgraph of a graph 𝐺, then that trail also exists in 𝐺. (Contributed by BTernaryTau, 22-Oct-2023.)
⊢ (𝑆 SubGraph 𝐺 → (𝐹(Trails‘𝑆)𝑃 → 𝐹(Trails‘𝐺)𝑃))
Theorem	subgrpth 35375	If a path exists in a subgraph of a graph 𝐺, then that path also exists in 𝐺. (Contributed by BTernaryTau, 22-Oct-2023.)
⊢ (𝑆 SubGraph 𝐺 → (𝐹(Paths‘𝑆)𝑃 → 𝐹(Paths‘𝐺)𝑃))
Theorem	subgrcycl 35376	If a cycle exists in a subgraph of a graph 𝐺, then that cycle also exists in 𝐺. (Contributed by BTernaryTau, 23-Oct-2023.)
⊢ (𝑆 SubGraph 𝐺 → (𝐹(Cycles‘𝑆)𝑃 → 𝐹(Cycles‘𝐺)𝑃))
Theorem	cusgr3cyclex 35377*	Every complete simple graph with more than two vertices has a 3-cycle. (Contributed by BTernaryTau, 4-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    ⇒   ⊢ ((𝐺 ∈ ComplUSGraph ∧ 2 < (♯‘𝑉)) → ∃𝑓∃𝑝(𝑓(Cycles‘𝐺)𝑝 ∧ (♯‘𝑓) = 3))
Theorem	loop1cycl 35378*	A hypergraph has a cycle of length one if and only if it has a loop. (Contributed by BTernaryTau, 13-Oct-2023.)
⊢ (𝐺 ∈ UHGraph → (∃𝑓∃𝑝(𝑓(Cycles‘𝐺)𝑝 ∧ (♯‘𝑓) = 1 ∧ (𝑝‘0) = 𝐴) ↔ {𝐴} ∈ (Edg‘𝐺)))
Theorem	2cycld 35379	Construction of a 2-cycle from two given edges in a graph. (Contributed by BTernaryTau, 16-Oct-2023.)
⊢ 𝑃 = ⟨“𝐴𝐵𝐶”⟩    &   ⊢ 𝐹 = ⟨“𝐽𝐾”⟩    &   ⊢ (𝜑 → (𝐴 ∈ 𝑉 ∧ 𝐵 ∈ 𝑉 ∧ 𝐶 ∈ 𝑉))    &   ⊢ (𝜑 → (𝐴 ≠ 𝐵 ∧ 𝐵 ≠ 𝐶))    &   ⊢ (𝜑 → ({𝐴, 𝐵} ⊆ (𝐼‘𝐽) ∧ {𝐵, 𝐶} ⊆ (𝐼‘𝐾)))    &   ⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐼 = (iEdg‘𝐺)    &   ⊢ (𝜑 → 𝐽 ≠ 𝐾)    &   ⊢ (𝜑 → 𝐴 = 𝐶)    ⇒   ⊢ (𝜑 → 𝐹(Cycles‘𝐺)𝑃)
Theorem	2cycl2d 35380	Construction of a 2-cycle from two given edges in a graph. (Contributed by BTernaryTau, 16-Oct-2023.)
⊢ 𝑃 = ⟨“𝐴𝐵𝐴”⟩    &   ⊢ 𝐹 = ⟨“𝐽𝐾”⟩    &   ⊢ (𝜑 → (𝐴 ∈ 𝑉 ∧ 𝐵 ∈ 𝑉))    &   ⊢ (𝜑 → 𝐴 ≠ 𝐵)    &   ⊢ (𝜑 → ({𝐴, 𝐵} ⊆ (𝐼‘𝐽) ∧ {𝐴, 𝐵} ⊆ (𝐼‘𝐾)))    &   ⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐼 = (iEdg‘𝐺)    &   ⊢ (𝜑 → 𝐽 ≠ 𝐾)    ⇒   ⊢ (𝜑 → 𝐹(Cycles‘𝐺)𝑃)
Theorem	umgr2cycllem 35381*	Lemma for umgr2cycl 35382. (Contributed by BTernaryTau, 17-Oct-2023.)
⊢ 𝐹 = ⟨“𝐽𝐾”⟩    &   ⊢ 𝐼 = (iEdg‘𝐺)    &   ⊢ (𝜑 → 𝐺 ∈ UMGraph)    &   ⊢ (𝜑 → 𝐽 ∈ dom 𝐼)    &   ⊢ (𝜑 → 𝐽 ≠ 𝐾)    &   ⊢ (𝜑 → (𝐼‘𝐽) = (𝐼‘𝐾))    ⇒   ⊢ (𝜑 → ∃𝑝 𝐹(Cycles‘𝐺)𝑝)
Theorem	umgr2cycl 35382*	A multigraph with two distinct edges that connect the same vertices has a 2-cycle. (Contributed by BTernaryTau, 17-Oct-2023.)
⊢ 𝐼 = (iEdg‘𝐺)    ⇒   ⊢ ((𝐺 ∈ UMGraph ∧ ∃𝑗 ∈ dom 𝐼∃𝑘 ∈ dom 𝐼((𝐼‘𝑗) = (𝐼‘𝑘) ∧ 𝑗 ≠ 𝑘)) → ∃𝑓∃𝑝(𝑓(Cycles‘𝐺)𝑝 ∧ (♯‘𝑓) = 2))
21.5.4.1  Acyclic graphs
Syntax	cacycgr 35383	Extend class notation with acyclic graphs.
class AcyclicGraph
Definition	df-acycgr 35384*	Define the class of all acyclic graphs. A graph is called acyclic if it has no (non-trivial) cycles. (Contributed by BTernaryTau, 11-Oct-2023.)
⊢ AcyclicGraph = {𝑔 ∣ ¬ ∃𝑓∃𝑝(𝑓(Cycles‘𝑔)𝑝 ∧ 𝑓 ≠ ∅)}
Theorem	dfacycgr1 35385*	An alternate definition of the class of all acyclic graphs that requires all cycles to be trivial. (Contributed by BTernaryTau, 11-Oct-2023.)
⊢ AcyclicGraph = {𝑔 ∣ ∀𝑓∀𝑝(𝑓(Cycles‘𝑔)𝑝 → 𝑓 = ∅)}
Theorem	isacycgr 35386*	The property of being an acyclic graph. (Contributed by BTernaryTau, 11-Oct-2023.)
⊢ (𝐺 ∈ 𝑊 → (𝐺 ∈ AcyclicGraph ↔ ¬ ∃𝑓∃𝑝(𝑓(Cycles‘𝐺)𝑝 ∧ 𝑓 ≠ ∅)))
Theorem	isacycgr1 35387*	The property of being an acyclic graph. (Contributed by BTernaryTau, 11-Oct-2023.)
⊢ (𝐺 ∈ 𝑊 → (𝐺 ∈ AcyclicGraph ↔ ∀𝑓∀𝑝(𝑓(Cycles‘𝐺)𝑝 → 𝑓 = ∅)))
Theorem	acycgrcycl 35388	Any cycle in an acyclic graph is trivial (i.e. has one vertex and no edges). (Contributed by BTernaryTau, 12-Oct-2023.)
⊢ ((𝐺 ∈ AcyclicGraph ∧ 𝐹(Cycles‘𝐺)𝑃) → 𝐹 = ∅)
Theorem	acycgr0v 35389	A null graph (with no vertices) is an acyclic graph. (Contributed by BTernaryTau, 11-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    ⇒   ⊢ ((𝐺 ∈ 𝑊 ∧ 𝑉 = ∅) → 𝐺 ∈ AcyclicGraph)
Theorem	acycgr1v 35390	A multigraph with one vertex is an acyclic graph. (Contributed by BTernaryTau, 12-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    ⇒   ⊢ ((𝐺 ∈ UMGraph ∧ (♯‘𝑉) = 1) → 𝐺 ∈ AcyclicGraph)
Theorem	acycgr2v 35391	A simple graph with two vertices is an acyclic graph. (Contributed by BTernaryTau, 12-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    ⇒   ⊢ ((𝐺 ∈ USGraph ∧ (♯‘𝑉) = 2) → 𝐺 ∈ AcyclicGraph)
Theorem	prclisacycgr 35392*	A proper class (representing a null graph, see vtxvalprc 29134) has the property of an acyclic graph (see also acycgr0v 35389). (Contributed by BTernaryTau, 11-Oct-2023.) (New usage is discouraged.)
⊢ 𝑉 = (Vtx‘𝐺)    ⇒   ⊢ (¬ 𝐺 ∈ V → ¬ ∃𝑓∃𝑝(𝑓(Cycles‘𝐺)𝑝 ∧ 𝑓 ≠ ∅))
Theorem	acycgrislfgr 35393*	An acyclic hypergraph is a loop-free hypergraph. (Contributed by BTernaryTau, 15-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    &   ⊢ 𝐼 = (iEdg‘𝐺)    ⇒   ⊢ ((𝐺 ∈ AcyclicGraph ∧ 𝐺 ∈ UHGraph) → 𝐼:dom 𝐼⟶{𝑥 ∈ 𝒫 𝑉 ∣ 2 ≤ (♯‘𝑥)})
Theorem	upgracycumgr 35394	An acyclic pseudograph is a multigraph. (Contributed by BTernaryTau, 15-Oct-2023.)
⊢ ((𝐺 ∈ UPGraph ∧ 𝐺 ∈ AcyclicGraph) → 𝐺 ∈ UMGraph)
Theorem	umgracycusgr 35395	An acyclic multigraph is a simple graph. (Contributed by BTernaryTau, 17-Oct-2023.)
⊢ ((𝐺 ∈ UMGraph ∧ 𝐺 ∈ AcyclicGraph) → 𝐺 ∈ USGraph)
Theorem	upgracycusgr 35396	An acyclic pseudograph is a simple graph. (Contributed by BTernaryTau, 17-Oct-2023.)
⊢ ((𝐺 ∈ UPGraph ∧ 𝐺 ∈ AcyclicGraph) → 𝐺 ∈ USGraph)
Theorem	cusgracyclt3v 35397	A complete simple graph is acyclic if and only if it has fewer than three vertices. (Contributed by BTernaryTau, 20-Oct-2023.)
⊢ 𝑉 = (Vtx‘𝐺)    ⇒   ⊢ (𝐺 ∈ ComplUSGraph → (𝐺 ∈ AcyclicGraph ↔ (♯‘𝑉) < 3))
Theorem	pthacycspth 35398	A path in an acyclic graph is a simple path. (Contributed by BTernaryTau, 21-Oct-2023.)
⊢ ((𝐺 ∈ AcyclicGraph ∧ 𝐹(Paths‘𝐺)𝑃) → 𝐹(SPaths‘𝐺)𝑃)
Theorem	acycgrsubgr 35399	The subgraph of an acyclic graph is also acyclic. (Contributed by BTernaryTau, 23-Oct-2023.)
⊢ ((𝐺 ∈ AcyclicGraph ∧ 𝑆 SubGraph 𝐺) → 𝑆 ∈ AcyclicGraph)
21.6  Mathbox for Mario Carneiro
21.6.1  Predicate calculus with all distinct variables
Axiom	ax-7d 35400*	Distinct variable version of ax-11 2170. (Contributed by Mario Carneiro, 14-Aug-2015.)
⊢ (∀𝑥∀𝑦𝜑 → ∀𝑦∀𝑥𝜑)