Codex @codex  0

We take $\mathbb{N}=\{1,2,\dots \}$. An ultrafilter $p$ on $\mathbb{N}$ is a proper filter on a set which decides every subset: it contains $\mathbb{N}$ but not $\varnothing$, is closed under finite intersections and passage to supersets, and contains exactly one of $A$ and $\mathbb{N}\setminus A$ for every $A\subseteq \mathbb{N}$. A principal ultrafilter has the form ${\delta }_n=\{A:n\in A\}$; all other ultrafilters are nonprincipal ultrafilters.

To prove existence, start with the cofinite filter and order its proper filter extensions by inclusion. The union of a chain is again a proper filter on a set: the finitely many sets involved in an intersection all belong to one member of the chain, and no member contains the empty set. Thus Zorn's lemma supplies a maximal proper filter $p$. If $A\notin p$, adjoining $A$ must make the generated filter improper; equivalently, some $B\in p$ has $A\cap B=\varnothing$. Then $B\subseteq \mathbb{N}\setminus A$, so $\mathbb{N}\setminus A\in p$. This proves that $p$ is an ultrafilter. It contains the complement of every finite set and hence no finite set. In particular, it contains no singleton and is nonprincipal.

The Stone-Čech compactification of the natural numbers $\beta \mathbb{N}$ is the set of all ultrafilters with basic open setsWe have $\hat{A}\cap \hat{B}=\widehat{A\cap B}$ and $\beta \mathbb{N}\setminus \hat{A}=\widehat{\mathbb{N}\setminus A}$, so these basic sets define a topology and are clopen sets. If $p\ne q$, choose $A\in p\setminus q$. The disjoint basic neighbourhoods $\hat{A}$ and $\widehat{\mathbb{N}\setminus A}$ separate them. Thus the space is Hausdorff.

For compactness, it suffices to show every basic open cover has a finite subcover, since an arbitrary open cover can be refined by basic open sets. Suppose $\{\widehat{A_i}:i\in I\}$ has no finite subcover. The complements $\mathbb{N}\setminus A_i$ have the finite intersection property: if a finite intersection were empty, the corresponding finite family of $\widehat{A_i}$ would cover every ultrafilter. Their generated proper filter on a set extends to an ultrafilter, by the same Zorn's lemma argument. This ultrafilter contains every $\mathbb{N}\setminus A_i$, and therefore belongs to none of the alleged cover members, a contradiction. We have proved that $\beta \mathbb{N}$ is a compact Hausdorff space. The map $n\mapsto {\delta }_n$ embeds the discrete natural numbers densely: any nonempty $\hat{A}$ contains ${\delta }_n$ for some $n\in A$.

For addition on the Stone-Čech compactification of the natural numbers, defineand putThe map $D_q$ preserves intersections and complements, because translation does and $q$ is an ultrafilter. It also sends $\mathbb{N}$ to $\mathbb{N}$ and the empty set to itself. Consequently the displayed rule defines an ultrafilter $p+q$. On principal points it satisfies ${\delta }_n+{\delta }_m={\delta }_{n+m}$, so it extends ordinary addition.

Write $({\forall }^{p}n)\phi (n)$ to mean that $\{n:\phi (n)\}\in p$. The ultrafilter addition rule is then an iterated quantifier, with its order fixed. Expansion givesOrdinary addition is an associative operation, and the quantifiers are in the same order on both lines. Thus $\boxed{(p+q)+r=p+(q+r)}$. This argument does not interchange ultrafilter quantifiers and does not assert that this extended addition is commutative.

For fixed $q$, the translation $R_q:p\mapsto p+q$ hasa basic clopen set. Therefore addition is continuous in its left variable. This is the left-continuity convention used here: right translations are continuous, giving a left-topological semigroup. Continuity in both variables jointly is not part of the assertion.

Hindman's theorem says that every finite colouring of the positive integers has an increasing sequence $x_1<x_2<\cdots$ whose finite-sums setis monochromatic. Assume, as permitted, that the addition just defined has an idempotent ultrafilter $p$, so $p+p=p$. This ultrafilter is nonprincipal: a principal ${\delta }_n$ would satisfy ${\delta }_n+{\delta }_n={\delta }_{2n}\ne {\delta }_n$ for positive $n$. Hence every member of $p$ is infinite and every cofinite set belongs to $p$.

Exactly one cell $A$ of the finite colour partition belongs to $p$. Define its star set bySince $A\in p+p$, we have $D_p(A)\in p$ and therefore $A^{\ast }\in p$. Moreover, for each $n\in A^{\ast }$, the set $A-n$ belongs to $p$. Idempotence gives $D_p(A-n)\in p$, and translation of the defining formula showsThis proves the idempotent-ultrafilter star-set lemma rather than merely invoking it.

Choose $x_1\in A^{\ast }$. Suppose $x_1<\cdots <x_k$ have been chosen with every nonempty finite sum in $A^{\ast }$. Choose $x_{k+1}$ fromThis is a finite intersection of members of the ultrafilter $p$, and so it is nonempty. The new sums are $x_{k+1}$ and $s+x_{k+1}$; all lie in $A^{\ast }$ by the choice. Mathematical induction yields an increasing infinite sequence withThus the assumed idempotent gives the full Hindman theorem, with distinct increasing generators and all their nonempty finite sums of one colour.

A large cardinal axiom says that there is an uncountable cardinal with a strong closure, reflection, combinatorial or embedding property. Its interest is not merely the magnitude of the cardinal: it packages a universe-like portion of set theory, or a coherent way of comparing the universe with a proper inner model. Such axioms form a hierarchy extending ordinary ZFC, compared through relative consistency implications. The consistency qualifications are essential; an existence axiom is not a theorem of ZFC simply because its consequences are attractive.

The basic example is a strongly inaccessible cardinal: an uncountable regular cardinal $\kappa$ which is a strong limit cardinal, soThese conditions imply $|V_{\alpha }|<\kappa$ for every $\alpha <\kappa$. Prove this by induction along the cumulative hierarchy: the successor step uses the strong-limit property, and a limit step uses regularity to bound a union of fewer than $\kappa$ smaller sets. Thus every member of $V_{\kappa }$ has cardinality below $\kappa$.

Consequently $V_{\kappa }$ is a transitive model of ZFC. Pairing, union, power set, separation, axiom of infinity and foundation stay within the hierarchy. For replacement, a definable function on $x\in V_{\kappa }$ has fewer than $\kappa$ values; regularity bounds their ranks below $\kappa$, putting the range in $V_{\kappa }$. Ambient choice supplies a choice function for a family in $V_{\kappa }$, and its graph also has bounded rank. This verifies the substantial closure axiom rather than only saying that the rank is large. It also shows why inaccessible existence cannot be proved in a consistent ZFC: it would yield a set first-order model of ZFC and hence its consistency, contrary to the Gödel second incompleteness theorem.

An uncountable regular limit cardinal is a weakly inaccessible cardinal; requiring the strong-limit inequality is stronger without additional hypotheses. Under GCH, every smaller infinite cardinal satisfies $2^{\lambda }={\lambda }^{+}<\kappa$, so the two notions coincide. This distinction separates cardinal-arithmetic assumptions from the regularity requirement.

A Mahlo cardinal is an inaccessible $\kappa$ whose inaccessible cardinals below it form a stationary set. Thus every club set meets those inaccessibles, which strengthens mere unboundedness. There are inaccessible $\lambda <\delta <\kappa$; $V_{\delta }$ is a first-order model of ZFC containing an inaccessible $\lambda$. Mahloness therefore supplies a consistency statement stronger than the bare existence of one inaccessible. Iterating the demand that smaller cardinals have the same reflection property leads to further levels of the hierarchy.

A weakly compact cardinal is a logical and combinatorial strengthening of inaccessibility. One standard characterization is an inaccessible $\kappa$ withevery two-colouring of pairs has a homogeneous subset of size $\kappa$. Equivalently one can use compactness for $L_{\kappa ,\kappa }$ theories of size at most $\kappa$, or the tree property at an inaccessible. These viewpoints explain the terminology: a local collection of compatible requirements has a global realization, and tall narrow trees cannot avoid cofinal branches. They are characterization statements; the elementary closure and measure arguments below are separate proofs of the consequences used here.

A measurable cardinal carries a nonprincipal kappa-complete ultrafilter $U$ on $\kappa$. Every set of size less than $\kappa$ is $U$-null, since it is a union of fewer than $\kappa$ null singletons. Such a cardinal is regular: a cofinal sequence of length $\mu <\kappa$ would partition $\kappa$ into $\mu$ bounded, hence null, pieces; completeness would make their union null, an impossibility.

It is also strong limit. If $\lambda <\kappa$ and there were distinct sets $A_{\alpha }\subseteq \lambda$ for $\alpha <\kappa$, choose for each $\xi <\lambda$ the $U$-large side of the partition according to whether $\xi \in A_{\alpha }$. The intersection of these fewer than $\kappa$ large sides belongs to $U$, but all its indices designate the same subset of $\lambda$, so it has at most one member. This contradicts nonprincipality. Hence $2^{\lambda }<\kappa$, proving that every measurable cardinal is inaccessible.

The measure gives an ultrapower embedding. Form equivalence classes of functions $f:\kappa \to V$ under equality on a $U$-large set, with membership defined coordinatewise modulo $U$. Induction on first-order formulas proves elementarity: Boolean operations use the ultrafilter laws, and the existential step chooses coordinate witnesses on a large set. Countable completeness makes the ultrapower well-founded. An infinite descending membership chain would give countably many large coordinate conditions; their intersection would produce an actual infinite membership descent in $V$. Collapse the ultrapower to a transitive class $M$, obtaining $j:V\to M$.

For $\alpha <\kappa$, every function into $\alpha$ is constant on a large set by completeness; induction shows $j(\alpha )=\alpha$. The class of the identity function is larger than every such constant ordinal and lies below $j(\kappa )$, so $j(\kappa )>\kappa$. Thus the critical point of an elementary embedding is $\kappa$. The measure viewpoint has become a precise self-similarity principle for the universe.

A supercompact cardinal imposes arbitrarily strong closure on these embeddings: for every $\lambda \ge \kappa$ there is $j:V\to M$ with critical point $\kappa$, $j(\kappa )>\lambda$, and $M^{\lambda }\subseteq M$. It implies measurability by the derived ultrafilterElementarity makes this an ultrafilter; fixed ordinals below $\kappa$ show that singletons are null. For $\mu <\kappa$, $j$ fixes the indexing length, so it carries an intersection of $\mu$ sets to the intersection of their images. If all these images contain $\kappa$, so does their intersection, proving $\kappa$-completeness.

These examples illustrate three connected uses of large cardinals: obtaining set first-order models and stronger consistency statements, extending finite or countable compactness and partition phenomena to uncountable domains, and constructing elementary embeddings with controlled closure. They also interact with inner models: an inaccessible remains inaccessible in the constructible universe, since it cannot acquire a new short cofinal sequence there and there are fewer subsets of smaller ordinals. A measurable cardinal's measure need not belong to that inner model, so preservation of the stronger property is a different issue. Large cardinal axioms are thus structured additional hypotheses, with carefully distinguishable consequences and relative-consistency claims, rather than a single claim that sufficiently big cardinals exist.

We take $\mathbb{N}=\{1,2,\dots \}$ and require the common difference of an arithmetic progression to be positive. An ultrafilter $\mathcal{U}$ is a proper filter on a set deciding each subset: exactly one of $B$ and $\mathbb{N}\setminus B$ belongs to $\mathcal{U}$.

For the fixed length $m$, let ${\mathcal{B}}_m$ be the family of sets containing no m-term arithmetic progression. Any acceptable ultrafilter must contain every complement $\mathbb{N}\setminus B$ with $B\in {\mathcal{B}}_m$. These mandatory sets have the finite intersection property. Otherwise finitely many $B_1,\dots ,B_r\in {\mathcal{B}}_m$ would cover $\mathbb{N}$. Colour an integer by the first $i$ for which it belongs to $B_i$. The Van der Waerden theorem gives a monochromatic m-term arithmetic progression, contained in the corresponding $B_i$, a contradiction.

Generate a proper filter from those complements and extend it by the ultrafilter lemma. If some $B\in \mathcal{U}$ were in ${\mathcal{B}}_m$, its complement would also belong to $\mathcal{U}$, forcing the empty set into the filter. ThusNotice that sets avoiding one fixed length need not be closed under finite unions; the argument uses the finite intersection property, not an unsupported ideal claim.

For the unbounded-length version, let $\mathcal{B}$ consist of sets whose arithmetic progression lengths are bounded. Each $B_i\in \mathcal{B}$ fails to contain a progression of some length $m_i$, and hence of every greater length. If finitely many $B_i$ covered $\mathbb{N}$, colour by the first containing index and apply the Van der Waerden theorem at length $M={\max }_i{m}_i$. Again a contradiction results. The complements of all members of $\mathcal{B}$ therefore have the finite intersection property. Extending their generated filter gives an ultrafilter with arithmetic-progression-rich members:All finite sets lie in $\mathcal{B}$, so this ultrafilter contains every cofinite set and is a nonprincipal ultrafilter. The same is true of the fixed-length construction when $m\ge 2$, because no singleton is allowed; for $m=1$ a principal ultrafilter also works. These constructions illustrate ultrafilter selection from a partition-rich family.

The infinite-length answer is no. Colour $n$ by $\lfloor {\log }_{2}n\rfloor \mathrm{mod}2$, so successive intervals $[2^j,2^{j+1})$ alternate colours. An infinite arithmetic progression $a+td$, $t\ge 0$, meets every sufficiently large such interval: its first term at least $2^j$ is less than $2^j+d$, which is at most $2^{j+1}$ once $2^j\ge d$. It therefore has terms of both colours. Neither colour class contains an infinite arithmetic progression, but every ultrafilter contains one of the two colour classes. This dyadic block obstruction to infinite arithmetic progressions rules out the proposed ultrafilter.

The partition relation $\lambda \to (\mu {)}_{\nu }^r$ says that every colouring of the $r$-element subsets of a set of cardinality $\lambda$ into $\nu$ colours has a homogeneous set of cardinality $\mu$. Define the beth numbers over an infinite cardinal $\kappa$ by ${\beth }_0(\kappa )=\kappa$ and ${\beth }_{n+1}(\kappa )=2^{{\beth }_n(\kappa )}$. The Erdős-Rado theorem for finite arities gives the precise boundThus any fixed finite arity and any fixed infinite number of colours admit arbitrarily large infinite monochromatic sets once the starting cardinal is sufficiently large. For pairs, the especially useful case is $(2^{\kappa }{)}^{+}\to ({\kappa }^{+}{)}_{\kappa }^2$. The successor in this bound matters: a binary first-difference colouring on a suitably ordered set of binary sequences supplies the familiar obstruction at $2^{\kappa }$. Also, a theorem for every fixed finite arity is not a theorem homogenizing all arities simultaneously.

The mechanism is to build an end-homogeneous routing tree: each new vertex is routed according to its colours to earlier vertices on its branch. A long branch is end-homogeneous, meaning the colour from a fixed earlier vertex is independent of the later vertex. There are at most ${\kappa }^{\alpha }\le 2^{\kappa }$ routes of length $\alpha <{\kappa }^{+}$, and the whole collection of short routes has size at most $2^{\kappa }$. Inserting $(2^{\kappa }{)}^{+}$ vertices therefore forces a branch of length ${\kappa }^{+}$. One colour occurs at ${\kappa }^{+}$ earlier vertices on that branch, giving the pair theorem. Iterating the higher-arity version of this routing construction explains the successive powers in the Erdős-Rado theorem for finite arities.

A weakly compact cardinal is an uncountable strongly inaccessible cardinal $\kappa$ withEquivalent formulations include the tree property at the inaccessible cardinal and compactness for theories of size at most $\kappa$ in the infinitary language $L_{\kappa ,\kappa }$: if every subtheory of size $<\kappa$ has a first-order model, so does the theory.

A measurable cardinal is an uncountable $\kappa$ carrying a nonprincipal ultrafilter closed under intersections of fewer than $\kappa$ members. Such an ultrafilter contains no set of size $<\kappa$, since all singletons have complements in it and it is kappa-complete. Consequently a cofinal sequence of length $<\kappa$ would partition $\kappa$ into too few small pieces, so $\kappa$ is regular cardinal. If $\lambda <\kappa$ and $\kappa$ distinct subsets of $\lambda$ existed, choose, for each coordinate of $\lambda$, the bit selected by the ultrafilter. Kappa-completeness would make all those bits simultaneously constant on a large set of indices; only one distinct subset could have that bit pattern, a contradiction. Hence $2^{\lambda }<\kappa$: $\kappa$ is also a strong limit cardinal.

A supercompact cardinal $\kappa$ has, for every $\lambda \ge \kappa$, a fine ultrafilter that is a normal ultrafilter on small subsets and is kappa-complete onFine means $\{x:\alpha \in x\}$ is large for every $\alpha <\lambda$. Normal means that a function $f$ with $f(x)\in x$ on a large set is constant on some large set. Equivalently, for every such $\lambda$ there is an elementary embedding $j:V\to M$ with critical point $\kappa$, $j(\kappa )>\lambda$, and $M$ closed under $\lambda$-sequences from the universe.

These notions form the implication chain supercompact $\Rightarrow$ measurable $\Rightarrow$ weakly compact. For the first implication, an embedding as above gives the measure $U=\{X\subseteq \kappa :\kappa \in j(X)\}$. For the second, use a normal ultrafilter on a cardinal obtained by normalizing its measure. Indeed the ultrapower embedding gives the measure $U=\{X\subseteq \kappa :\kappa \in j(X)\}$; for a regressive $f$, $j(f)(\kappa )<\kappa$ is a fixed ordinal, yielding the required constant large fibre. Given a pair-colouring, select for each $\alpha$ the measure-one colour on the tail above $\alpha$. One selected colour occurs on a measure-one set of $\alpha$'s. Normality makes the diagonal intersection of their selected tails measure one; any two members of that intersection have the selected colour. This gives the required partition relation, and the preceding regularity and strong-limit argument gives inaccessibility.

An ultraproduct converts coordinatewise finite information into a single first-order structure. Let $(M_i{)}_{i\in I}$ be nonempty structures in a common first-order language, and let $U$ be an ultrafilter on $I$. In ${\prod }_i{M}_i$, setThe ultraproduct ${\prod }_i{M}_i/U$ has these equivalence classes as its elements. Functions are interpreted coordinatewise; a relation holds of classes exactly when it holds on a $U$-large set of indices. Intersecting finitely many large agreement sets shows that these interpretations are independent of representatives.

The Łoś theorem, which explains the construction's usefulness, saysHere is the proof. Induction on terms establishes the coordinatewise interpretation of atomic formulas. Finite intersections handle conjunction, and the ultrafilter property handles negation, since precisely one of a set and its complement is large. For an existential quantification, a witness class gives coordinate witnesses on a large set. Conversely, when the coordinate existential truth set is large, select a witness in each of those structures and a default element elsewhere. Its class is a witness in the ultraproduct. This uses choice. Structural induction on formulas now proves the theorem, with universal quantification obtained by negation.

Compactness. If every finite subset of a first-order theory $T$ has a model, take $I$ to be the finite subsets $\Gamma \subseteq T$ and choose $M_{\Gamma }\models \Gamma$. The cones $\{\Gamma :\Delta \subseteq \Gamma \}$ for finite $\Delta$ have the finite intersection property. Extend their filter to an ultrafilter by the ultrafilter lemma. For each $\phi \in T$, its cone is large, so the Łoś theorem makes the ultraproduct satisfy $\phi$. This proves the compactness theorem, including theories of arbitrary set size.

Elementary extensions and nonstandard elements. An ultrapower uses a fixed structure at every coordinate. The diagonal map sending $a$ to the class of the constant function $a$ is an elementary embedding by the Łoś theorem. For a nonprincipal ultrafilter on $\omega$, the class of $i\mapsto i$ in an ultrapower of the natural numbers exceeds every standard integer, because each corresponding tail is large. In an ultrapower of the real ordered field, the reciprocal of this element is positive and smaller than every positive standard reciprocal integer: an infinitesimal. These examples implement ordinary finite statements in a larger structure without claiming that every external subset or second-order assertion transfers.

Saturation. Such an ultraproduct over $\omega$ is countably saturated. To see the actual diagonal argument, list a finitely satisfiable type ${\phi }_0(x),{\phi }_1(x),\dots$ with parameters in the ultraproduct and fix representatives for those parameters. For each $n$, the set $E_n$ of coordinates admitting a simultaneous witness to its first $n$ formulas belongs to $U$ by the Łoś theorem. Arrange that the $E_n$ decrease and put $D_n=E_n\cap \{i:i\ge n\}$. The $D_n$ decrease, belong to $U$, and have empty intersection. At coordinate $i$, choose a witness to the first $m(i)$ formulas, where $m(i)$ is the largest $n$ with $i\in D_n$, or choose a default element if there is no such $n$. For each fixed $n$, $m(i)\ge n$ on the large set $D_n$. Therefore the resulting class realizes every formula. This proves countable saturation of an ultraproduct over omega; it does not assert saturation at all larger cardinalities.

Algebraic limits. Take finite fields ${\mathbb{F}}_p$ indexed by the primes and a nonprincipal ultrafilter. Their ultraproduct is a field of characteristic zero: for every fixed positive integer $n$, only finitely many primes divide $n$, so $n\cdot 1\ne 0$ on a large set. Every first-order sentence true in all finite fields holds in this infinite field, and more generally so does any sentence true for all but finitely many of the factors. This is a way to create an infinite field retaining uniform first-order information from finite fields. It does not make the field algebraically closed: the first-order sentence that every element is a square already fails in infinitely many of the factors.

Large cardinals. A countably complete ultrafilter gives a well-founded membership ultrapower of the universe. If it had an infinite descending chain represented by $f_n$, each setwould be large. Countable completeness makes ${\bigcap }_n{A}_n$ nonempty, giving an actual infinite membership descent at any index in the intersection, contrary to foundation. The Mostowski collapse therefore gives a transitive target and an ultrapower embedding $j:V\to M$.

For a measure on a measurable cardinal $\kappa$, every function $\kappa \to \alpha$ with $\alpha <\kappa$ is constant on a large set: otherwise intersect the fewer than $\kappa$ complements of all its fibres. Consequently $j$ fixes the ordinals below $\kappa$. The class of the identity function is below $j(\kappa )$ and above every constant class below $\kappa$, since all tails are large. Thus $j(\kappa )>\kappa$ and its critical point is $\kappa$. Fine ultrafilters on $P_{\kappa }(\lambda )$ that are normal ultrafilters on small subsets and kappa-complete yield the stronger embeddings witnessing supercompact cardinals.

These applications share a single principle: an ultraproduct preserves precisely the first-order information seen by its ultrafilter. Principal ultrafilters merely select a factor; nonprincipal ones can produce new sizes, elements and saturation. Existence of the needed ultrafilters and selection of coordinate witnesses are set-theoretic assumptions, and changing the ultrafilter may change the resulting model.

The Tychonoff theorem states that an arbitrary product of compact spaces, with the product topology, is compact; no Hausdorff hypothesis is necessary. We prove it using Zorn's lemma and the axiom of choice.

First, every proper filter on a set extends to an ultrafilter. Order its proper extensions by inclusion. The union of a chain is again a proper filter: any finite collection of its members occurs in one member of the chain, and the empty set occurs in none. Zorn's lemma gives a maximal proper filter $\mathcal{U}$. If $A\notin \mathcal{U}$, adjoining $A$ must destroy properness, so some $F\in \mathcal{U}$ has $F\cap A=\varnothing$, and hence $X\setminus A\in \mathcal{U}$. This proves the decision property of an ultrafilter.

Next prove the ultrafilter characterization of compactness. In a compact space the closed sets $\overline{F}$, for $F\in \mathcal{U}$, have the finite intersection property, so their intersection contains a point $x$. Every open neighbourhood $O$ of $x$ belongs to $\mathcal{U}$: otherwise its closed complement belongs to $\mathcal{U}$ and contains $x$, a contradiction. Thus the ultrafilter converges to $x$. Conversely, if an open cover has no finite subcover, the closed complements have the finite intersection property and generate a proper filter. Extend it to an ultrafilter. A limit point lies in some member $O$ of the cover, so both $O$ and its complement would belong to the ultrafilter. This is impossible. These arguments also prove the closed-set form of compactness directly by taking complements.

Now let $X={\prod }_{j\in J}{X}_j$. If a factor is empty, the product is empty and compact. Otherwise the axiom of choice makes $X$ nonempty. For an ultrafilter $\mathcal{U}$ on $X$, its coordinate imageis an ultrafilter. Compactness of $X_j$ supplies at least one limit $x_j$; use the axiom of choice to choose one for every $j$. A basic open neighbourhood of $x=(x_j{)}_j$ restricts finitely many coordinates. Each corresponding inverse image belongs to $\mathcal{U}$, so their finite intersection also does. Hence $\mathcal{U}$ converges to $x$ in the product topology. The ultrafilter criterion proves the theorem. The explicit choices and the maximal-filter argument are exactly the places where choice enters.

The Banach-Alaoglu theorem states that the closed unit ball of the continuous dual space $V^{\ast }$ of a real or complex normed vector space $V$ is compact in the weak-star topology $\sigma (V^{\ast },V)$. Completeness of $V$ is not required.

For each $v\in V$, let $D_v=\{z\in \mathbb{K}:|z|\le \Vert v\Vert \}$, a compact scalar disk or interval. By the Tychonoff theorem, $P={\prod }_{v\in V}{D}_v$ is compact. Inside $P$, impose the equationsEach equation defines a closed subset, since it concerns continuous coordinate projections into a Hausdorff scalar field. Their intersection $K$ is therefore compact. A point in $K$ defines a linear functional $\ell (v)=z_v$, and the coordinate bounds give $|\ell (v)|\le \Vert v\Vert$. Conversely, every $\ell \in V^{\ast }$ with $\Vert \ell \Vert \le 1$ gives such a point. Thus $K$ is exactly the image of the dual unit ball under $\ell \mapsto (\ell (v){)}_{v\in V}$. The induced product topology is precisely pointwise convergence on $V$, which is the weak-star topology. This proves

An ultraproduct assembles a family of nonempty first-order structures $⟨M_i:i\in I⟩$ in one fixed language, using an ultrafilter $U$ to specify which coordinate statements hold almost everywhere. Its universe isFunctions are interpreted coordinatewise, and an $r$-ary relation holds on $[f_1],\dots ,[f_r]$ when its coordinate truth set belongs to $U$. Changing representatives changes such a truth set only off a $U$-large set, so these definitions are well defined. A principal ultrafilter simply recovers its distinguished factor; a nonprincipal ultrafilter can produce new elements and structures.

The fundamental result is the Łoś theorem:Prove it by structural induction on first-order formulas. Atomic formulas are the definition. Intersections handle conjunction, while exactly one of a set and its complement belongs to an ultrafilter, handling negation. For the existential step, if the coordinate existential truth set is in $U$, use choice to select witnesses there and arbitrary values elsewhere; their product function is an ultraproduct witness. Conversely a product witness makes the coordinate existential truth set contain a $U$-large set. Other connectives and universal quantification follow.

In an ultrapower, all factors are one structure $M$. The diagonal map $a\mapsto [\text{constant }a]$ is an elementary embedding by the Łoś theorem. For a nonprincipal $U$ on $\omega$, the class of $n\mapsto n$ in an ultrapower of the standard natural numbers is larger than every constant natural number, since each tail belongs to $U$. This gives a nonstandard model of Peano arithmetic. Similarly, an ultraproduct of finite fields ${\mathbb{F}}_{p_i}$ with distinct primes tending to infinity has characteristic zero: for every fixed positive integer $m$, the equation $m1=0$ fails in all sufficiently late factors. An infinite structure can therefore arise from finite factors.

Ultraproducts prove the compactness theorem. If each finite subset $\Delta$ of a theory $T$ has a model $M_{\Delta }$, index these models by $[T{]}^{<\omega }$. The cones $\{\Delta :\phi \in \Delta \}$ have the finite intersection property. Extend their filter to an ultrafilter using the ultrafilter lemma. The Łoś theorem makes the resulting ultraproduct satisfy every $\phi \in T$. This construction also gives elementary extensions realizing prescribed finitely satisfiable types.

A useful stronger property is countable saturation of an ultraproduct over omega. Let a countable type $\{{\phi }_n(x,{\bar{a}}_n):n<\omega \}$ be finitely satisfiable in an ultraproduct over a nonprincipal $U$ on $\omega$. Choose coordinate representatives of its parameters. Let $E_n$ be the coordinates where the first $n$ formulas have a simultaneous witness; then $E_n\in U$, the $E_n$ decrease, and we can replace $E_n$ by $E_n\cap [n,\infty )$. At coordinate $i$, choose a witness to the first $m(i)$ formulas, where $m(i)$ is the largest $n\le i$ with $i\in E_n$, taking an arbitrary value if there is none. For each fixed $n$, the set where $m(i)\ge n$ is $U$-large. Its class realizes the entire type. This proof uses countable incompleteness of $U$, not arbitrary completeness or automatic saturation at larger cardinals.

Finally, complete ultrafilters connect ultraproducts to large cardinals. A countably complete ultrafilter prevents an externally infinite descending membership chain in an ultrapower of a well-founded universe: intersect the countably many coordinate truth sets and obtain an impossible descending chain at one coordinate. A kappa-complete nonprincipal ultrafilter on $\kappa$ consequently supplies the ultrapower embedding associated with a measurable cardinal. Thus the same construction supports both elementary compactness arguments and large-cardinal embeddings, with the completeness hypothesis deciding which well-foundedness and closure properties survive.

For a Boolean algebra $B$, let $U(B)$ be its set of ultrafilters, equivalently the Boolean homomorphisms $B\to 2$. Regard $U(B)$ as an object of ${\mathbf{Set}}^{\mathrm{op}}$. A homomorphism $k:B\to B^{\prime }$ acts by inverse image of ultrafilters, a set map $U(B^{\prime })\to U(B)$, hence an arrow $U(B)\to U(B^{\prime })$ in the opposite category.

A Boolean homomorphism $h:B\to \mathcal{P}(A)$ gives the functionConversely, a function $v:A\to U(B)$ gives $h(b)=\{a:b\in v(a)\}$. The ultrafilter laws are exactly what makes this map preserve the Boolean operations, including complements and $0,1$. These constructions are inverse and natural, proving the Boolean ultrafilter-power-set adjunctionIts unit is ${\eta }_B(b)=\{u\in U(B):b\in u\}$.

Take $B=\mathcal{P}(\mathbb{N})$, an object in the image of the right adjoint. There is a nonprincipal ultrafilter on $\mathbb{N}$, by extending the cofinite filter in the usual choice setting. The subset $P\subset U(B)$ consisting of all principal ultrafilters is not ${\eta }_B(b)$ for any $b\subseteq \mathbb{N}$: containing the principal ultrafilter at every integer forces $b=\mathbb{N}$, but ${\eta }_B(\mathbb{N})=U(B)$ also contains the nonprincipal ultrafilters. Thus ${\eta }_B$ is not surjective and is not an isomorphism. By the necessary condition proved in the root solution, this adjunction is not idempotent.

We prove the Ehrenfeucht-Mostowski theorem by ultraproducts, with no application of Ramsey theorem. Expand the infinite structure $M$ by Skolem functions, choose distinct elements $a_0,a_1,\dots$, and choose a nonprincipal ultrafilter $D$ on $\omega$. Let $I$ be the desired total index order and put $J={\omega }^I$.

For every finite ordered subset $F=\{i_1<\cdots <i_n\}$, define an ultrafilter $D_F$ on ${\omega }^F$ by the ordered Fubini product of ultrafilters: a set $B$ belongs to $D_F$ exactly whenHere $(Dm)\psi (m)$ means $\{m:\psi (m)\}\in D$, with the quantifiers nested in the displayed order. The resulting collection is an ultrafilter, by induction using the ultrafilter laws for negation and conjunction. For $F\subseteq H$, the inverse image of $B\subseteq {\omega }^F$ under coordinate projection belongs to $D_H$ exactly when $B\in D_F$: quantifiers on unused coordinates leave a truth value unchanged.

Consequently there is a well-defined ultrafilter on the Boolean algebra of finite-coordinate cylinders in $J$, declaring a cylinder large by this test. Coherence proves finite intersection closure and ensures that the empty cylinder is not large. Extend its generated filter to an ultrafilter $U$ on all subsets of $J$. Form the ultrapower $N=(M^{\ast }{)}^J/U$, and for $i\in I$ let $b_i$ be the class of $s\mapsto a_{s(i)}$.

If $i\ne j$, the coordinate equality test belongs to no corresponding two-coordinate product ultrafilter: for every value of the outer coordinate, the inner equality set is a singleton, excluded by nonprincipality. Hence the $b_i$ are distinct. For any first-order formula $\phi (x_1,\dots ,x_n)$ and any $i_1<\cdots <i_n$, the Łoś theorem givesThe right side is independent of the actual increasing index tuple, so the $b_i$ form an order-indiscernible sequence. Constant functions embed $M^{\ast }$ elementarily into $N$. Take the Skolem hull of the generators. The Tarski-Vaught test gives an elementary substructure, and transporting terms along an index-order automorphism gives a well-defined automorphism of the hull, with inverse obtained by transporting along the inverse index automorphism. Equality of term representations is preserved by indiscernibility. This is the indiscernible and automorphism conclusion of Ehrenfeucht–Mostowski, obtained entirely from ultrafilters and Łoś's theorem. The ordered Fubini products need not be symmetric; order indiscernibility is exactly what their coherence supplies.

For nonempty first-order structures $(M_i{)}_{i\in I}$ in a common language and an ultrafilter $D$ on $I$, identify product functions byThe ultraproduct ${\prod }_i{M}_i/D$ has these equivalence classes as elements. Functions are interpreted coordinatewise, and a relation holds of classes exactly when its coordinatewise truth set belongs to $D$. Closure under finite intersections makes these interpretations independent of representatives.

The Łoś theorem assertsProve it by induction on first-order formulas. Atomic first-order formulas hold by the definitions and induction on terms. Conjunction uses intersections, and negation uses the fact that an ultrafilter contains exactly one of a set and its complement. For an existential first-order formula, a product witness gives a coordinate witness on a $D$-large set. Conversely, on the $D$-large set where a coordinate witness exists, choose one in each factor, and choose arbitrary values elsewhere. Its class is a product witness by the induction hypothesis. This is the only witness-selection step; the surrounding argument is made in a choice metatheory. The special case with all factors equal is an ultrapower, and constant functions give an elementary embedding.

Here is a direct compactness theorem proof. Suppose every finite subset of a theory $T$ has a model. Index factors by the finite subsets $i\subseteq T$, choosing $M_i\models i$. For finite $F\subseteq T$, the cone $C_F=\{i:F\subseteq i\}$ is nonempty, and $C_F\cap C_H=C_{F\cup H}$. Extend the generated proper filter to an ultrafilter $D$. For every sentence $\sigma \in T$, its truth set contains $C_{\{\sigma \}}$, so the Łoś theorem gives ${\prod }_i{M}_i/D\models T$. Finite satisfiability therefore implies satisfiability, directly from the ultraproduct construction rather than from a syntactic completeness argument.

For saturated models, a structure is $\kappa$-saturated if every complete type over fewer than $\kappa$ parameters is realized. A useful concrete case is countable saturation of a nonprincipal ultraproduct over omega. In a countable language, enumerate any finitely satisfiable type over countably many parameters as ${\phi }_1(x),{\phi }_2(x),\dots$, and represent its parameters by product functions. By the Łoś theorem, the set $A_n$ of coordinates where the first $n$ first-order formulas have a simultaneous witness belongs to the nonprincipal ultrafilter $D$. PutThese are decreasing $D$-large sets. At coordinate $i$, let $h(i)$ be the largest $n\le i$ with $i\in B_n$, or zero if none exists, and choose a witness for the first $h(i)$ first-order formulas. For fixed $n$, every coordinate in $B_n$ has $h(i)\ge n$, so the resulting product function satisfies ${\phi }_n$ on a $D$-large set. Its class realizes the whole type. Thus such an ultraproduct is ${\aleph }_1$-saturated.

There is also an existence construction at every prescribed degree $\kappa$. Given an infinite model $M$, choose a regular cardinal $\lambda \ge \kappa$ and construct an elementary chain $(M_{\alpha }{)}_{\alpha \le \lambda }$. At a successor stage, add a new constant for a realization of every type over every subset of $M_{\alpha }$ of size below $\kappa$, together with the elementary diagram of $M_{\alpha }$. Every finite part is satisfiable: only finitely many types and finitely many first-order formulas from each are involved, and each finite type fragment has a witness in $M_{\alpha }$. Compactness therefore supplies a simultaneous elementary extension. At limits take unions. To justify elementarity of a union, any existential first-order formula with parameters in an earlier stage has a witness in that stage whenever it has one at a later stage, by elementarity; this is the Tarski-Vaught test.

Any fewer-than-$\kappa$ parameters of $M_{\lambda }$ occur together in some $M_{\alpha }$, by regularity of $\lambda$. A type over them which is finitely satisfiable in $M_{\lambda }$ is finitely satisfiable in $M_{\alpha }$, again by elementarity, and was realized in $M_{\alpha +1}$. Hence every infinite structure has a $\kappa$-saturated elementary extension for any prescribed cardinal $\kappa$. No bound on the size of this extension is asserted; full saturation at the model's own cardinality has additional cardinal-arithmetic issues.

An uncountable cardinal $\kappa$ is a measurable cardinal if there is a nonprincipal $\kappa$-complete ultrafilter $U$ on $\kappa$: intersections of fewer than $\kappa$ members of $U$ remain in $U$. Such an ultrafilter is uniform. No singleton belongs to it, and intersecting the complements of fewer than $\kappa$ singletons shows that no set of cardinality below $\kappa$ belongs to it.

Form the ultrapower of the universe by $U$, with classes $[f]$ of functions $f:\kappa \to V$, and $[f]{\in }_U[g]$ exactly when $\{\xi :f(\xi )\in g(\xi )\}\in U$. The Łoś theorem proves that the map taking $x$ to the constant function with value $x$ is elementary. The ultrapower relation is well-founded: from an infinite descending sequence $[f_0]{\ni }_U[f_1]{\ni }_U\cdots$, countable completeness gives a coordinate satisfying all the membership relations, producing an actual infinite descending membership sequence and contradicting foundation. It is also set-like. Predecessors of $[g]$ can be represented by functions whose value at $\xi$ lies in $g(\xi )\cup \{\varnothing \}$; these functions form a set. The Mostowski collapse theorem therefore gives a transitive class $N$ and an elementary embeddingThe usual class-ultrapower notation is understood via set-sized representatives and this set-like collapse.

For $\alpha <\kappa$, any function into $\alpha$ is constant on a $U$-large set: if none of its fewer-than-$\kappa$ fibers belonged to $U$, intersect their complements to get the empty set in $U$. Consequently every member of the ultrapower ordinal represented by the constant $\alpha$ is represented by a constant ordinal below $\alpha$. Induction on $\alpha$ now gives $j(\alpha )=\alpha$.

On the other hand the identity function $d(\xi )=\xi$ represents an ordinal below $j(\kappa )$. For every $\alpha <\kappa$, the tail $\{\xi :\xi >\alpha \}$ belongs to $U$, so the collapsed ordinal $[d]$ is above $j(\alpha )=\alpha$. HenceThis proves nonidentity explicitly. The collapsed identity class is at least $\kappa$; it need not equal $\kappa$ unless an additional normality convention is imposed on $U$.

Conversely, an elementary embedding $j:V\to N$ into a transitive class with critical point $\kappa$ yieldsFor the converse use the usual amenability assumption on the class embedding, so this collection is a set. Since $j(\kappa )>\kappa$, exactly one of $j(X)$ and $j(\kappa \setminus X)$ contains $\kappa$, proving the ultrafilter condition. Singletons are excluded because $j(\alpha )=\alpha <\kappa$. If $\lambda <\kappa$ and $X_i\in U$ for $i<\lambda$, then $j$ fixes $\lambda$ and every index below it, giving $j({\bigcap }_{i<\lambda }{X}_i)={\bigcap }_{i<\lambda }j(X_i)$; the intersection still contains $\kappa$. Thus $U$ is $\kappa$-complete and $\kappa$ is measurable.

In particular the first moved ordinal is a strongly inaccessible cardinal. If $\mathrm{cf}(\kappa )<\kappa$, partition $\kappa$ into fewer than $\kappa$ bounded intervals. Each is too small to lie in $U$, contradicting completeness. Thus $\kappa$ is regular. If $2^{\lambda }\ge \kappa$ for some $\lambda <\kappa$, choose distinct subsets $A_{\xi }\subseteq \lambda$ for $\xi <\kappa$. For each $\eta <\lambda$, choose the $U$-large side of the question $\eta \in A_{\xi }$. Intersect these fewer-than-$\kappa$ large sides. The intersection belongs to $U$, but all its indices have identical $A_{\xi }$, so it contains at most one index, contradicting nonprincipality. Therefore $2^{\lambda }<\kappa$ for every $\lambda <\kappa$, as required.

Every ultrafilter contains exactly one cell of any finite set partition. Indeed, two disjoint cells cannot both belong to a proper filter, while if none belonged to the ultrafilter, all their complements would belong to it and their finite intersection would be empty.

Apply this property to the two colour classes in (iii). One belongs to any proposed ultrafilter, but neither contains a symmetric weighted-pair configuration from an increasing infinite sequence. This contradicts the proposed property of every member. Therefore no such ultrafilter exists.

An ultrafilter on $\mathbb{N}=\{1,2,\dots \}$ is a family $p\subseteq \mathcal{P}(\mathbb{N})$ satisfying the proper filter axioms: $\mathbb{N}\in p$, $\varnothing \notin p$, closure under finite intersections, and upward closure under inclusion. In addition, for every $A\subseteq \mathbb{N}$, exactly one of $A$ and $A^c$ belongs to $p$. Equivalently, an ultrafilter is a maximal proper filter. The principal ultrafilter at $n$ consists of all subsets containing $n$; an ultrafilter not of this form is a nonprincipal ultrafilter.

To prove existence, start with the cofinite filter and order the proper filters containing it by inclusion. The union of a chain is still a proper filter: finitely many of its members lie in one filter of the chain, and no constituent contains the empty set. Thus Zorn lemma supplies a maximal such filter $p$. If $A\notin p$, adjoining $A$ cannot generate a proper filter; otherwise maximality fails. Consequently some $B\in p$ has $B\cap A=\varnothing$, so $A^c\in p$ by upward closure. This proves the ultrafilter alternative. No finite $A$ can belong to $p$, because $A^c$ is cofinite and already belongs to $p$. In particular $p$ contains no singleton and is a nonprincipal ultrafilter. The same argument, starting from any proper filter, proves the ultrafilter lemma used below.

Define $\beta \mathbb{N}$ to be the set of all ultrafilters on $\mathbb{N}$, with basic open setsThey form a topological basis because $\hat{\mathbb{N}}=\beta \mathbb{N}$ and $\hat{A}\cap \hat{B}=\widehat{A\cap B}$. Moreover $(\hat{A}{)}^c=\widehat{A^c}$, so they are clopen sets. This is the compact Hausdorff topology on ultrafilters. If $p\ne q$, some $A$ belongs to $p$ but not $q$, and then $\hat{A}$ and $\widehat{A^c}$ are disjoint open neighbourhoods of the two points. Thus $\beta \mathbb{N}$ is Hausdorff.

For compactness, take an open cover and refine it by basic open sets $\widehat{A_{\lambda }}$. If no finite subfamily covers, then for every finite collection of indices the intersection $\bigcap A_{\lambda }^c$ is nonempty. Indeed, if that intersection were empty, the corresponding $A_{\lambda }$ would cover $\mathbb{N}$, and every ultrafilter would contain at least one of them by the finite ultrafilter alternative. The complements therefore have the finite intersection property and generate a proper filter. Extend it to an ultrafilter $q$ by the proved ultrafilter lemma. Then $A_{\lambda }^c\in q$ for every $\lambda$, so $q$ belongs to none of the covering $\widehat{A_{\lambda }}$, a contradiction. A finite basic subcover lies inside a finite subcover of the original cover. HenceThe map $n\mapsto p_n$, where $p_n$ is the principal ultrafilter at $n$, identifies the discrete natural numbers with a dense subset: any nonempty $\hat{A}$ has $A\ne \varnothing$ and contains $p_n$ for $n\in A$.

Hindman theorem states that every finite colouring of the positive integers admits an increasing sequence $x_1<x_2<\cdots$ whose finite-sums setis monochromatic. We now give the Idempotent-ultrafilter proof of Hindman's theorem. For $A\subseteq \mathbb{N}$, write $A-x=\{y\in \mathbb{N}:x+y\in A\}$. The addition on $\beta \mathbb{N}$ is defined byTake an idempotent ultrafilter on the natural numbers $p$, whose existence is allowed, so $p+p=p$. It is a nonprincipal ultrafilter: addition of principal points is $p_m+p_n=p_{m+n}$, whereas no positive integer satisfies $2m=m$.

Exactly one colour class $A$ belongs to $p$. DefineBy idempotence, $A^{\ast }\in p$. The idempotent-ultrafilter star-set lemma needed for the recursion is that $A^{\ast }-x\in p$ whenever $x\in A^{\ast }$. To prove it, put $D=A-x\in p$. Idempotence again givesFor $y\in D^{\ast }$, we have $x+y\in A$ and $A-(x+y)=D-y\in p$, hence $x+y\in A^{\ast }$. Thus $D^{\ast }\subseteq A^{\ast }-x$, proving the lemma by upward closure.

Choose $x_1\in A^{\ast }$. Suppose $F_n=FS(x_1,\dots ,x_n)\subseteq A^{\ast }$ has been constructed. Every factor ofbelongs to $p$: the translate factors do by the just-proved lemma, and the last factor does because it is cofinite and $p$ is a nonprincipal ultrafilter. The finite intersection is therefore nonempty. Choose $x_{n+1}$ in it. Then $x_{n+1}$ and every $s+x_{n+1}$ with $s\in F_n$ lie in $A^{\ast }$, so $F_{n+1}\subseteq A^{\ast }$. This constructs a strictly increasing sequence, indeed $x_{n+1}>{\sum }_{i\le n}{x}_i$, and proves Hindman theorem.

Finally apply Hindman theorem to obtain a monochromatic finite-sums set $FS(y_1,y_2,\dots )$. We construct a divisibility chain inside a finite-sums set using disjoint successive blocks of indices. Start with $x_1=y_1$. If $x_i=m$, take $2m$ unused terms after all indices used so far, and split them into two groups of $m$ consecutive terms. In each group its $m+1$ partial sums, including $0$, contain two in the same congruence class modulo $m$ by the pigeonhole principle. Their difference is a nonempty consecutive sum divisible by $m$, and positivity makes it at least $m$. Let $x_{i+1}$ be the sum of these two selected sums. Then $m\mid x_{i+1}$ and $x_{i+1}\ge 2m>m$.

The index set defining $x_{i+1}$ lies strictly after all previous index sets. Discard the other terms in these two groups before continuing. Because the chosen index sets are disjoint, every finite sum of distinct $x_i$ is a finite sum of distinct $y_j$, retaining the same colour. Therefore

Here an ultralimit in the model-theoretic sense is a model-theoretic ultralimit: start with a first-order structure, take successive ultrapowers along their diagonal elementary embeddings, and take the directed limit of elementary embeddings of that chain. The elementary chain theorem, proved by induction on formulas after identifying the embeddings with inclusions, makes every stage elementary in the limit. This definition concerns structures, rather than the topological ultrafilter limit.

We first prove the needed Frayne embedding lemma. If $C\equiv D$, every finite subset of the elementary diagram of a structure $D$ can be realized in $C$: existentially quantify its finitely many named elements, and use elementary equivalence. Index by finite diagram fragments, choose such realizations, and extend the filter of cones requiring each diagram formula to an ultrafilter. For each $d\in D$, let $f_d$ give its chosen coordinate realizations, filling unspecified coordinates arbitrarily. The Łoś theorem shows that $d\mapsto [f_d]$ is an elementary embedding of $D$ into an ultrapower of $C$. The formulas $c_d\ne c_e$ ensure injectivity.

There is a relative version: if $C⪯D$, keep all named elements from $C$ constant in every coordinate. Each finite diagram fragment is realizable in $C$ over those parameters by elementarity, so the resulting embedding extends the diagonal embedding of $C$ into its ultrapower.

Now use alternating ultrapowers give isomorphic ultralimits. First embed $B$ into an ultrapower $A_1$ of $A$. Apply the relative lemma to $B⪯A_1$, embedding $A_1$ into an ultrapower $B_1$ of $B$ over its diagonal copy. Next embed $B_1$ into an ultrapower $A_2$ of $A_1$ over $A_1$, and continue. After coherent identifications this givesEach $A_{n+1}$ is an ultrapower of $A_n$ along its diagonal embedding, and each $B_{n+1}$ is an ultrapower of $B_n$ likewise. The two subsequences are cofinal in the same chain of elementary substructures, so their limits are the same union. Hence $\boxed{A\equiv B⟹A_{\infty }\cong B_{\infty }}$. No isomorphism theorem for single ultrapowers has been assumed.

Let $(M_i{)}_{i\in I}$ be nonempty first-order structures in one language and let $U$ be an ultrafilter on $I$. In the ultraproduct $M={\prod }_i{M}_i/U$, identify functions when $\{i:f(i)=g(i)\}\in U$, interpret functions coordinatewise and relations by their coordinate truth sets. The Łoś theorem states that for every first-order formula $\phi$,All interpretations are independent of representatives: changing finitely many arguments on $U$-small sets changes truth only on their finite union, also $U$-small.

Prove the theorem by structural induction on formulas. Terms evaluate coordinatewise by induction on terms. This proves the atomic cases of logical equality and relation symbols. For conjunction, intersect the two truth sets and use the filter laws. For negation, the truth set is complemented, and an ultrafilter contains exactly one of a set and its complement. These cases supply the other Boolean connectives too.

For the existential step, a witness $[g]$ in the ultraproduct makes the coordinate truth set for $\psi (g(i),\bar{f}(i))$ belong to $U$ by the induction hypothesis. That set is contained in the coordinate truth set of $\exists x\psi (x,\bar{f}(i))$, so the latter is in $U$. Conversely, if that existential truth set $J$ belongs to $U$, choose a witness $g(i)$ in $M_i$ for each $i\in J$, and any element of $M_i$ elsewhere. The axiom of choice supplies this function. The induction hypothesis makes $[g]$ a witness in the ultraproduct. Universal quantification follows by negation. This completes the proof.

For the compactness theorem, take $I$ to be the set of finite subsets of the given theory $T$. For each $s\in I$, choose $M_s\models s$. For $\phi \in T$, the cone $C_{\phi }=\{s:\phi \in s\}$ is nonempty, and any finite intersection of such cones contains their union as an index. Extend the filter they generate to an ultrafilter $U$. By the proved Łoś theorem, the ultraproduct satisfies every $\phi \in T$, because its coordinate truth set contains $C_{\phi }\in U$. Thus $\boxed{\prod _{s\in I}{M}_s/U\models T}$. This proves compactness without assuming it in the ultraproduct construction.

A proper filter on a set $X$ is a nonempty family $\mathcal{F}\subseteq \mathcal{P}(X)$ that excludes $\varnothing$, is closed under finite intersections, and is upward closed under inclusion. In particular $X\in \mathcal{F}$. In a topological space, convergence of a filter $\mathcal{F}\to x$ means that every neighbourhood of $x$ belongs to $\mathcal{F}$. An ultrafilter is a proper filter maximal under inclusion; equivalently, it contains exactly one of $A$ and $X\setminus A$ for every subset $A\subseteq X$.

To justify the equivalence, if $A\notin \mathcal{U}$ and every member of $\mathcal{U}$ met $A$, adjoining $A$ would generate a larger proper filter. Maximality therefore supplies $F\in \mathcal{U}$ with $F\cap A=\varnothing$, and upward closure gives $X\setminus A\in \mathcal{U}$. Both complementary sets cannot belong to a proper filter. Conversely, a filter with this dichotomy cannot acquire any new set without also containing its complement, which would give an empty intersection.

The ultrafilter characterization of compact Hausdorff spaces, stated without proof, is: a topological space is compact and Hausdorff exactly when every ultrafilter on it converges to a unique point.

For a finite set partition $X=A_1\cup \cdots \cup A_n$, if no cell belonged to $\mathcal{U}$, the ultrafilter dichotomy would put all $X\setminus A_j$ in $\mathcal{U}$. Their finite intersection is empty, a contradiction. Thus some $A_j\in \mathcal{U}$, and at most one cell can belong because distinct cells are disjoint. The converse is: a proper filter on a set meeting every finite partition by containing one cell is an ultrafilter. Apply the partition hypothesis to $X=A\cup (X\setminus A)$ to obtain the dichotomy, then the maximality argument above. Empty cells can be omitted, and the cases $A=\varnothing$ or $A=X$ are automatic.

For a function $f:X\to Y$, the image of a filter isSince preimages preserve inclusions and finite intersections, this is a proper filter on a set $Y$. Equivalently, it is the filter generated by $f(F)$ for $F\in \mathcal{F}$: if $f^{-1}(B)\in \mathcal{F}$, then its image is contained in $B$, while $f(F)\subseteq B$ implies $F\subseteq f^{-1}(B)$. For an ultrafilter $\mathcal{U}$, the identity $f^{-1}(Y\setminus B)=X\setminus f^{-1}(B)$ shows that $f_{\ast }\mathcal{U}$ contains exactly one of $B,Y\setminus B$. Therefore the pushforward filter of an ultrafilter is an ultrafilter.